One-time purchase · Yours forever · Works with the free Graphnosis download
These packs contain zero actual data. Graphnosis has zero servers.
A pack is a structure: engram schemas, recall patterns, and connector guides. Your memory is
built from your own files, on your own device. Nothing is pre-loaded, nothing is shared,
and no Graphnosis server ever touches it. Every kit is a starting point — fully expandable
and customizable by you, without limitation, under the free or any tiered plan.
95 packs
MemoryStudioStarter Kit
Engineering
Software Developer
For engineers who build with AI daily
Architecture recall — decision rationale and alternatives surfaced before any code change
Debug history — prior attempts for similar errors recalled automatically
Review context — codebase decisions attached to every code review session
Post-fix capture — lessons from resolved issues saved after each fix
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Every pack on this page includes the same contents listed above.
For reliability engineers managing complex systems
Incident history — fault signatures and prior resolution paths recalled before triage
Service dependency recall — known integration quirks surfaced before changes
Post-incident capture — root cause and timeline saved for future pattern matching
Configuration drift memory — prior deviations and their consequences recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers building and maintaining data pipelines
Schema decision recall — why a column was typed, renamed, or deprecated
Pipeline rule memory — transformation logic and its exceptions recalled on edit
Data quality history — prior anomalies, their sources, and resolutions surfaced
Upstream dependency recall — known issues with source systems, before you rely on them
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers building, training, and shipping machine learning systems
Experiment memory — dataset choices, hyperparameter rationale, and what failed recalled before the next run
Architecture decision recall — why a model architecture was chosen or abandoned, surfaced before redesign
Deployment lessons — production failures, rollback reasons, and tuning outcomes recalled on next deploy
Evaluation history — benchmark results and their caveats recalled before claiming improvement
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers doing threat modeling, vulnerability research, and incident response
Threat model history — prior threat assessments recalled before designing new surfaces
Vulnerability investigation memory — prior research threads, false positives, and confirmed findings recalled
Incident pattern recall — past incident signatures and response playbooks surfaced on new alerts
Tool and control decision recall — why a specific control was chosen or rejected, before revisiting
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers working on firmware, microcontrollers, and hardware-software interfaces
Firmware revision history — what changed, why, and what broke recalled before touching the same subsystem
Hardware-software interface memory — pin assignments, timing constraints, and interface quirks recalled on revisit
Debug pattern recall — prior fault signatures and their root causes surfaced before deep-dive sessions
Supplier and component recall — known errata, substitution decisions, and qualification outcomes by part
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers designing, analyzing, and iterating physical systems
Design decision recall — tolerance rationale, material selection, and trade-off history surfaced before revision
Simulation and FEA memory — what was modeled, what assumptions were made, and what was learned
Supplier and manufacturing recall — known process constraints and supplier quirks recalled before spec release
Failure mode history — prior field failures and their root causes recalled before similar designs are released
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
EMC and signal integrity memory — prior failures, layout changes, and test outcomes recalled before respins
BOM and supplier recall — last-time-buy decisions, approved alternates, and lead time history per component
Test and bring-up history — what failed on first power-on, what the fix was, recalled before new board revision
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers managing long-lifecycle infrastructure projects
Design intent recall — why a structural decision was made recalled across handoffs, inspections, and repairs
Inspection and condition history — prior findings, deferred items, and remediation outcomes per structure
Regulatory and code memory — variance decisions, agency correspondence, and interpretation history recalled
Project constraint memory — site conditions, client constraints, and value-engineering trade-offs recalled on revisit
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers in process industries: oil & gas, pharma, chemicals, specialty materials
HAZOP and MOC memory — prior hazard studies, management-of-change decisions, and deferred actions recalled
Process parameter history — operating envelope rationale, alarm set-point decisions, and deviation log recalled
Scale-up decision recall — what worked in the lab, what failed at scale, and why — surfaced before the next scale-up
Regulatory and PSM history — OSHA 1910.119 action items, PHA findings, and compliance rationale recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers managing requirements, interfaces, and trade studies across complex programs
Requirement traceability memory — where a requirement came from, how it changed, and what drove each revision
Trade study history — what was compared, what was traded, and why the chosen architecture won
Interface control recall — ICD versions, open interface issues, and resolution history recalled across teams
Verification rationale — why a test method was chosen, what was waived, and the justification recalled on audit
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For operators and engineers managing critical plant assets
Equipment fault history — prior failure signatures recalled before maintenance decisions
Maintenance lineage — what was done, when, by whom, and what happened after
Deviation tracking — procedure non-compliances and their outcomes, recalled on next trigger
Failure pattern recognition — recurring precursor conditions surfaced across time
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For quality engineers in regulated manufacturing environments
Inspection history — prior non-conformances recalled before each audit or review
Procedure deviation memory — past deviations and corrective actions surfaced on demand
Supplier quality recall — known issues with materials and vendors, before ordering
Corrective action lineage — what was approved, what was closed, what recurred
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Site history recall — what was found, what was done, what was deferred at this location
Equipment quirk memory — known workarounds and vendor advisories recalled before work
Safety incident recall — near-misses and prior incidents surfaced before similar tasks
Handoff capture — end-of-shift notes saved for the next technician automatically
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For control room operators and engineers in power, water, and gas utilities
Alarm and event history — SCADA event logs ingested via OPC-UA or Modbus webhook; recurring alarm patterns recalled before shift handoff
Switching and isolation memory — prior procedure sequences, interlock rationale, and near-miss history recalled before issuing switching orders
Asset configuration recall — protection relay settings, set-point history, and non-standard configurations recalled by asset ID
Regulatory and NERC CIP recall — compliance action items, audit findings, and corrective measures surfaced before inspections
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For line supervisors and process engineers on production floors
PLC alarm memory — alarm logs ingested via OPC-UA or Modbus webhook; recurring fault signatures recalled before production decisions
Changeover and setup recall — prior setup parameters, tooling notes, and first-article outcomes recalled per product or SKU
Line operator tribal knowledge — informal workarounds and machine quirks captured and recalled across crew rotations
Quality excursion history — prior OOS events, their root causes, and what containment worked recalled before similar runs
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For facilities managers and building engineers overseeing complex properties
BMS event log memory — fault and alarm history ingested via BACnet/Modbus webhook; recurring patterns recalled per system or zone
Vendor and contractor recall — prior work quality, response time, pricing history, and known issues recalled before dispatching
Preventive maintenance memory — what was inspected, what was deferred, and what recurred recalled per asset
Tenant and occupant context — prior complaints, accommodations, and space history recalled before service calls
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For lab managers and scientists operating HPLC, mass spec, microscopy, and other instruments
Instrument run history — serial/USB output ingested directly; calibration drift, OOS results, and anomalies recalled per instrument
Calibration and qualification memory — IQ/OQ/PQ records, standard performance baselines, and requalification triggers recalled
Protocol deviation log — prior deviations, their impact, and corrective actions recalled before repeat runs
Method and SOP recall — which method version was used, what changed between versions, and why recalled on audit
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For engineers operating robot fleets, autonomous systems, and mechatronic platforms
Fleet telemetry memory — ROS topic data and sensor logs ingested via webhook; failure mode patterns recalled per robot or platform
Calibration and tuning history — what parameters were changed, when, why, and the resulting performance recalled per unit
Firmware and software trail — prior versions, rollback reasons, and field-observed regressions recalled before updates
Field condition memory — site-specific environmental factors, surface anomalies, and obstacle patterns recalled before deployment
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For upstream and midstream engineers managing wells, pipelines, and production facilities
Well and asset history — completion records, workover rationale, and production anomalies recalled per well; ingests via webhook or file
ESP and artificial lift recall — prior failure patterns, run-life history, and pump selection rationale recalled before interventions
HSE incident memory — near-misses, root causes, and corrective action outcomes recalled before similar operations
Subsurface knowledge recall — geological interpretations, reservoir behavior notes, and petrophysical decisions recalled per field
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For HVAC engineers and commissioning agents managing building system performance
Control sequence memory — BACnet/Modbus webhook ingest; sequence-of-operations rationale and override history recalled per AHU or zone
BAS performance history — seasonal baselines, trending anomalies, and energy deviation patterns recalled before recommissioning
Commissioning issue log — prior deficiencies, their resolution status, and recurring failures recalled across project phases
Vendor and service recall — prior service work, warranty claims, and known equipment quirks recalled before maintenance decisions
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For aircraft maintenance engineers and MRO technicians
Tech log and defect history — prior write-ups, deferred defects, and MEL history recalled per tail number before each inspection
Recurring squawk memory — chronic defect patterns and prior troubleshooting paths recalled before similar fault investigations
Part and component recall — AD compliance status, known service difficulties, and substitution history recalled per P/N
Task card and procedure memory — prior execution notes, deviations, and lessons captured for future maintainers via op-log
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For developers building agents that accumulate knowledge over time
Cross-run recall — prior findings surfaced before each new search run, no duplicate effort
Source provenance memory — where each finding came from, recalled on synthesis
Hypothesis lineage — what was tested, what was abandoned, why
Post-run capture — new findings saved after each run for future recall
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For agents that handle ongoing customer or partner relationships
Contact history recall — prior interactions, commitments, and open issues before each run
Account context — relationship health signals and account history recalled automatically
Commitment tracking — what was promised, to whom, when — surfaced on next contact
Post-interaction capture — outcome and next actions saved after each session
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For AI agents that act on codebases across multiple runs
Architecture context recall — prior design decisions recalled before each coding run
Debug history — prior attempts and their outcomes recalled when a similar error recurs
Change rationale memory — why a function was written this way, recalled on modification
Post-run capture — new decisions and lessons saved after each agent session
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For teams managing multi-year projects with complex decision trails
Decision archaeology — why was this chosen? Rationale, alternatives, and approvals recalled
Lessons learned recall — prior project outcomes surfaced before the same choice arises
Stakeholder history — prior positions, commitments, and relationships recalled
Risk history — what was flagged, what materialised, and what was dismissed last time
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For organizations preserving knowledge through staff transitions
Succession context — what the departing person knew, recalled for the replacement
Tribal knowledge capture — undocumented practices saved before they walk out the door
Org decision history — how policies were formed and why, recalled when challenged
Relationship handoff memory — client and partner history passed to the next account holder
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For researchers and innovation teams spanning years of work
Hypothesis lineage — what was tested, what was rejected, what was shelved and why
Experiment thread memory — prior results recalled before designing the next experiment
Collaboration history — who contributed what, when, and what they found
Dead-end recall — what has already been tried and failed, before the team tries it again
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For care coordinators managing multi-provider patient journeys
Multi-provider context recall — all prior touchpoints surfaced before each handoff
Referral history — what was referred, what was received back, what was lost
Care plan memory — current plan and prior versions recalled when reviewing
Patient preference recall — documented preferences and prior communications surfaced
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For clinical researchers managing protocols and trial data
Protocol version recall — what changed, when, why, and what was the IRB response
Adverse event memory — prior events, their resolution, and reporting history
Subject history recall — prior visit data and protocol deviations surfaced per subject
Regulatory correspondence memory — FDA/EMA interactions recalled before submissions
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For lawyers who handle ongoing client relationships and matters
Client history recall — prior commitments, open issues, and relationship context before each call
Matter memory — prior advice, positions taken, and key dates recalled on demand
Commitment tracking — verbal and written commitments recalled by client and matter
Prior work recall — what was drafted, negotiated, or decided in prior engagements
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Deposition history — prior witness statements and inconsistencies recalled
Exhibit and document recall — where a document came from and why it matters, surfaced on demand
Opposing counsel pattern memory — prior positions and tactics recalled before proceedings
Timeline construction — key events and their sources recalled when building arguments
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For compliance counsel managing regulatory relationships
Regulatory correspondence recall — prior interactions with agencies surfaced before submissions
Audit history memory — what was found, what was remediated, what recurred
Enforcement pattern recall — prior actions and their outcomes, recalled when a new matter opens
Compliance program memory — what controls exist, why, and what changed them
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For IP counsel managing prosecution and portfolio strategy
Prior art recall — what has already been found and where, before a new search
Claim evolution memory — how claims changed during prosecution and why
Prosecution history recall — examiner positions, responses, and outcomes surfaced
Portfolio gap analysis — what technology areas are covered vs. exposed, recalled on demand
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For counsel protecting confidential business information
Asset inventory recall — what is designated as trade secret, who has access, and when
Disclosure history — every authorized disclosure recalled before a new one is considered
Access log recall — prior access events and their context surfaced in disputes
NDA and agreement memory — prior confidentiality terms recalled before new negotiations
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For counsel and advisors running M&A and licensing IP reviews
Portfolio assessment memory — prior validity opinions and red flags recalled per asset
Licensing history recall — prior term negotiations and outcomes surfaced in new deals
Freedom-to-operate recall — prior FTO analyses recalled before new product launches
Chain of title memory — ownership history and transfer records surfaced on demand
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For analysts and portfolio managers building long-term theses
Thesis history — prior investment theses, conviction changes, and outcomes recalled
Management context recall — prior conversations, commitments, and patterns surfaced
Sector knowledge memory — prior analysis and its conclusions recalled on new coverage
Contrarian signal capture — notes saved when conviction diverges from consensus
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Loss event recall — what happened, root cause, and reserve history for each event
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For advisors managing long-term client financial relationships
Account history recall — prior positions, decisions, and rationale before each review
Meeting memory — prior conversations, commitments, and concerns surfaced before calls
Life event recall — milestones and their financial implications remembered across years
Goal evolution memory — how client objectives changed and why, recalled on planning
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For policy professionals managing complex regulatory dockets
Regulatory history recall — prior agency positions and their evolution surfaced
Stakeholder position memory — who said what, when, and how their view shifted
Consultation record recall — prior public comment rounds and their outcomes
Policy decision lineage — what was decided, who approved it, and what changed it
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For records officers and FOIA coordinators managing disclosure workflows
Request history recall — prior requests, their scope, and processing decisions surfaced
Redaction rationale memory — why a specific exemption was applied, recalled in appeals
Requester history recall — prior contacts and their patterns surfaced per requester
Disclosure precedent recall — what was released in prior similar requests
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For program managers overseeing grants and funded initiatives
Award history recall — prior grant terms, conditions, and monitoring outcomes
Grantee context memory — compliance history and prior findings recalled per recipient
Reporting and milestone recall — what was due, what was submitted, what was flagged
Program design memory — prior implementation decisions and their outcomes recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Government & Public Sector ·Grant & Program Management
For export control officers and ITAR-regulated contractors
Authorization history recall — prior licenses, their scope, and expiry recalled before transactions
Jurisdiction memory — prior commodity jurisdiction determinations and their rationale
Disclosure and access tracking — what was shared with whom, recalled in audits
Empowered Official memory — prior compliance decisions and their basis surfaced on demand
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For program managers on DoD contracts (DFARS / CMMC)
Contract requirement recall — CDRL, SOW, and clause obligations surfaced before deliverables
Deliverable history — what was submitted, what was rejected, and the resolution recalled
DCSA and CMMC context memory — prior assessment findings and remediation history
Subcontractor flow-down recall — which clauses were passed down, to whom, and when
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For systems engineers on long-lifecycle aerospace and defense programs
Specification history recall — what changed between baselines, why, and what drove it
Test history memory — what was tested, what failed, and the resolution for each anomaly
Configuration item recall — current configuration and its change history surfaced on demand
Interface decision memory — why a specific interface was defined this way, recalled on modification
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For authors protecting and developing long-form creative work
Story bible recall — characters, rules, timelines, and world-building recalled on demand
Draft lineage — prior versions, what changed, and why — never lost between sessions
Continuity memory — prior established facts recalled before writing new scenes
Unpublished work boundary — Sensitive tier engaged; zero creative content reaches cloud AI
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For designers protecting unreleased visual and product work
Asset provenance recall — when each element was created, its source, and its status
Iteration history — what changed between versions and the brief behind each change
Style guide memory — brand rules and their rationale recalled across all sessions
IP boundary enforcement — unreleased assets stay local; no cloud ingestion
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For composers, producers, and studios protecting unreleased work
Composition history — what was written, when, and in what context, recalled on demand
Sample provenance recall — what was used, from where, and its clearance status
Rights and ownership memory — who owns what, what was licensed, and on what terms
Unreleased work boundary — zero audio content or metadata reaches external services
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For journalists managing ongoing source relationships
Source history recall — prior interactions, reliability signals, and relationship context
Contact commitment memory — what a source said they would provide, recalled on follow-up
Source pattern recall — prior inconsistencies or reliability issues surfaced before use
Sensitive source protection — Sensitive tier; source identities never reach cloud AI
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For investigative journalists tracking complex multi-thread stories
Evidence chain recall — what was found, where, and what it connects to, surfaced on demand
Timeline memory — the sequence of events and their sources recalled when writing
Entity relationship recall — how people, organizations, and documents connect, across time
Dead end memory — what was investigated and cleared, to avoid repeating the work
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For researchers maintaining an auditable experimental record
Experiment lineage — prior protocols, results, and deviations recalled before each run
Reagent and material recall — lot numbers, supplier issues, and substitutions surfaced
Hypothesis history — what was tested, what was abandoned, and the evidence trail
Protocol version memory — what changed between versions and the rationale
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For researchers managing large bodies of literature over time
Prior synthesis recall — what you concluded from a paper, recalled years later
Research thread memory — the line of inquiry and its evolution across time
Citation relationship recall — how sources connect to each other and to your argument
Gap memory — open questions and their prior investigation history surfaced on demand
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For researchers managing collaborations, grants, and IRB obligations
Collaboration history — who contributed what, on what timeline, and with what constraints
Data provenance recall — where each dataset came from and its usage restrictions
IRB and ethics memory — prior approvals, amendments, and their conditions recalled
Grant obligation recall — reporting deadlines, deliverables, and milestones surfaced
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For doctoral researchers and academics managing long research arcs
Literature synthesis memory — what you concluded from prior reading, recalled on demand
Dissertation thread recall — the argument, its evolution, and committee feedback over time
Advisor and supervisor memory — prior feedback, commitments, and agreed directions
Conference and submission history — what was presented, rejected, or accepted, and why
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For L&D professionals managing learning programs across an organization
Program history recall — what was run, how it performed, and what changed it
Competency gap memory — prior needs analyses and their recommended interventions
Learner cohort history — prior engagement patterns recalled when designing new programs
Vendor and content history — prior assessments of learning vendors and content quality
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For educators who teach, advise, research, and develop curriculum across years
Course and lesson memory — what worked, what fell flat, and why recalled before the next semester
Student and cohort patterns — recurring misconceptions, advising notes, and feedback themes by topic
Curriculum evolution recall — why content changed, what was cut, and what was added and when
Research-teaching balance — open questions from the classroom that fed research, and vice versa, recalled across projects
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For field technicians serving equipment at customer sites
Site history recall — what was found, what was done, and what was deferred at this location
Equipment quirk memory — known workarounds and vendor advisories recalled before work
Customer commitment recall — what was promised at prior visits, surfaced on return
Handoff capture — end-of-visit notes saved for the next technician automatically
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For safety officers and HSE professionals in operational environments
Incident history recall — prior incidents with similar characteristics surfaced before work
Near-miss memory — what was reported, investigated, and remediated recalled on demand
Procedure deviation recall — past deviations and their consequences surfaced before sign-off
Inspection history memory — prior findings recalled before each new audit cycle
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For fleet managers overseeing vehicles, drivers, and mobile assets
Vehicle maintenance history — recurring faults, repair outcomes, and deferred items recalled per asset
Driver and operator notes — performance patterns, incident history, and coaching context recalled before reviews
Vendor and repair shop recall — reliability, pricing history, and turnaround patterns by provider
Route and deployment memory — operational constraints, seasonal issues, and site-specific conditions recalled before dispatch
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For architects managing complex projects through design and approvals
Design decision recall — why a specific approach was taken and who approved it
Code compliance memory — prior interpretations and rulings recalled before submissions
Client brief history — how requirements evolved and what was agreed at each stage
Value engineering recall — prior cost decisions and their consequences surfaced on new projects
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
RFI history recall — prior questions, answers, and the decisions they produced
Change order memory — what was requested, what was approved, and at what cost
Subcontractor relationship recall — prior performance, issues, and commitments surfaced
Punch list and defect memory — open items and their resolution history across the project
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Architecture & Construction ·Construction Project Management
For architects and developers in feasibility, programming, and early-phase research
Site investigation memory — zoning history, environmental findings, and regulatory constraints recalled per site
Program and brief evolution — how client requirements shifted and why, recalled before design decisions crystallize
Precedent and typology recall — prior case studies, their lessons, and how they influenced the design direction
Consultant and stakeholder context — who said what in early-phase meetings, recalled before design reviews
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Architecture & Construction ·Pre-Design & Site Research
For account executives managing complex B2B relationships
Account history recall — full relationship context before every call or meeting
Risk signal memory — health indicators and churn signals recalled across the relationship
Renewal context recall — account history synthesised before renewal conversations
Objection history — how prior objections were handled and what the outcome was
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For CSMs managing ongoing customer health and expansion
Health signal recall — prior engagement patterns and risk indicators surfaced on review
Expansion opportunity memory — prior conversations about adjacent needs recalled
Escalation history — prior issues, their resolution, and the relationship impact recalled
QBR and review memory — prior commitments made in business reviews surfaced at renewal
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For operations leads managing supplier relationships and contracts
Supplier history recall — prior performance, issues, and negotiation outcomes surfaced
Contract term memory — SLA obligations, renewal dates, and key clauses recalled
Incident and SLA history — prior failures, credits, and remediation recalled per vendor
Procurement decision recall — why a vendor was chosen over alternatives, surfaced on review
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For founders and CEOs managing strategy, board relationships, and organizational context
Strategic decision memory — why a direction was chosen, what was rejected, and the reasoning recalled before pivots
Board and investor context — prior positions, sensitivities, and commitment history recalled before each meeting
Stakeholder and partner recall — relationship history, key agreements, and open items recalled before conversations
Organizational decision trail — hiring decisions, structure changes, and cultural commitments recalled across time
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For COOs and heads of operations driving cross-functional execution
Process decision memory — why an operational process was designed a certain way, recalled before redesign
Cross-functional context — what each team committed to, what slipped, and why recalled before planning cycles
Vendor and infrastructure recall — prior performance, contract terms, and negotiation outcomes per supplier
OKR and initiative history — prior goal rationale, missed targets and their causes, surfaced before the next cycle
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Business Operations & B2B ·COO & Operations Leader
For CFOs and finance leaders managing planning, reporting, and investor relationships
Financial decision memory — model assumptions, scenario rationale, and forecast variance explanations recalled
Investor and lender context — prior conversations, commitments, covenants, and sensitivities recalled before calls
Audit and compliance history — prior findings, management responses, and remediation status recalled per cycle
Board reporting recall — prior narratives, questions raised, and action items committed to recalled before prep
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For CMOs and heads of marketing managing brand, campaigns, and agency relationships
Brand decision memory — positioning rationale, rejected directions, and message evolution recalled before campaigns
Campaign history — prior performance, audience response, channel decisions, and lessons recalled before planning
Agency and partner recall — prior briefs, feedback patterns, and relationship context recalled before reviews
Competitive intelligence — prior analysis, positioning shifts, and response decisions recalled as market evolves
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For chiefs of staff and senior EAs supporting principals at the executive level
Principal context — preferences, working style, prior decisions, and standing commitments recalled before each engagement
Stakeholder relationship memory — who matters to the principal, prior interactions, and open threads recalled before prep
Meeting and initiative history — what was agreed, what was deferred, and what is still open recalled before follow-ups
Recurring brief memory — board prep, investor calls, leadership offsites — prior versions and their evolution recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Board and leadership narrative history — prior strategy presentations, objections raised, and direction given recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Business Operations & B2B ·Strategy & Corporate Development
For HRBPs embedded in business units, supporting leaders and employees
Manager and leader context — prior coaching conversations, recurring issues, and commitments recalled before 1:1s
Employee relations history — prior concerns raised, informal resolutions, and patterns recalled across the business unit
Organizational change memory — restructuring rationale, role changes, and transition decisions recalled on revisit
Talent and performance recall — prior review themes, development conversations, and succession context per employee
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For recruiters and talent acquisition leads managing pipelines and hiring decisions
Hiring manager context — role requirements, past feedback patterns, and preference history recalled before sourcing
Candidate history — prior conversations, prior applications, and reasons for past decisions recalled per candidate
Sourcing decision memory — which channels worked for which roles, and why alternatives were deprioritized
Offer and negotiation recall — prior ranges, counteroffers, and outcome patterns recalled before new offers
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
HR & People Operations ·Talent Acquisition & Recruiting
For people ops teams managing onboarding, policies, and the employee lifecycle
Onboarding memory — what each new hire was told, what was deferred, and what follow-ups are open per person
Policy decision trail — why a policy was written as it was, prior interpretations, and edge cases recalled on update
Offboarding and transition recall — prior handoff quality, knowledge transfer gaps, and lessons for next time
Systems and tooling history — prior vendor decisions, migration rationale, and known issues recalled before changes
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For HR leads managing investigations, policy compliance, and workplace matters
Investigation and case history — prior similar matters, their outcomes, and precedent recalled before new cases
Policy interpretation memory — prior edge cases, legal guidance received, and decisions made recalled on similar facts
Disciplinary and performance action trail — prior warnings, PIPs, and outcomes recalled when similar situations recur
Regulatory and audit history — prior agency interactions, findings, and corrective actions recalled before compliance reviews
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
HR & People Operations ·Employee Relations & HR Compliance
For total rewards leaders managing comp philosophy, benchmarking, and benefits programs
Compensation philosophy memory — prior band decisions, equity refresh rationale, and exception history recalled before cycles
Benchmarking and survey history — which surveys were used, what adjustments were made, and why recalled before renewal
Benefits design decision trail — prior vendor selection, plan design changes, and employee feedback recalled before open enrollment
Executive comp and equity recall — grant history, vesting decisions, and board approval context recalled per individual
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For HR leaders and finance partners doing headcount planning and org design
Headcount decision memory — prior approvals, rejected requests, and rationale recalled before planning cycles
Org design history — prior structures considered, trade-offs made, and outcomes recalled before redesign
Skills gap and capability recall — prior assessments, build-buy-partner decisions, and their outcomes by function
Attrition and retention memory — exit themes, prior retention interventions, and their effectiveness recalled
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For executive directors, program officers, and development staff
Funder memory — grant history, program officer preferences, prior rejections and approvals recalled per funder
Program outcome recall — impact data, beneficiary stories, and metrics surfaced for reports and renewals
Board and stakeholder context — member priorities, prior decisions, and relationship history recalled before meetings
Donor relationship memory — giving history, communications, stewardship commitments, and recognition recalled per donor
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Non-Profits & Communities ·Non-Profit & Foundation
For associations, faith communities, unions, and large member-based organizations
Member history recall — engagement patterns, needs, volunteer roles, and prior interactions recalled per member or household
Program and event memory — attendance history, feedback, recurring issues, and lessons-learned surfaced per program cycle
Committee and leadership context — working group decisions, term history, conflict resolutions, and institutional knowledge recalled
Partnership and coalition memory — MOU terms, partner contacts, collaboration history, and shared-goal alignment recalled before outreach
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Non-Profits & Communities ·Large Community & Membership Org
For volunteer coordinators managing large, diverse, rotating volunteer pools
Volunteer history recall — skills, availability, prior roles, and reliability patterns recalled before scheduling
Onboarding and training memory — what each volunteer has completed, what was waived, and what needs renewal
Recognition and engagement recall — milestones, preferences, and prior appreciation history recalled before outreach
Role and event memory — which volunteers worked which events, how they performed, and what they requested
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For healthcare compliance officers managing PHI controls and audit readiness
Policy version history — current and superseded HIPAA policies recalled before any update or audit
Training completion memory — staff training records, attestations, and gaps recalled before audits or enforcement events
Breach log recall — prior incidents, investigation outcomes, and OCR correspondence recalled before new assessments
Business Associate Agreement tracking — BAA status, covered entities, and data flows recalled before vendor onboarding
Risk assessment trail — prior risk analyses, mitigation decisions, and residual risk rationale recalled before reviews
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For DPOs and privacy teams managing GDPR obligations across jurisdictions
ROPA memory — Records of Processing Activities, lawful bases, and retention schedules recalled before assessments
DPIA trail — Data Protection Impact Assessments, outcomes, and residual risk decisions recalled before new processing activities
DSR handling history — prior Data Subject Requests, response timelines, and exceptions recalled before new requests
Supervisory authority correspondence — prior DPA interactions, enforcement actions, and commitments recalled before filings
Third-country transfer records — SCCs, adequacy decisions, and TIA outcomes recalled before new vendor contracts
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For internal audit and controls teams managing Sarbanes-Oxley compliance programs
Control inventory recall — control descriptions, owners, and test results recalled before walkthrough preparation
Deficiency tracking — prior control deficiencies, remediation commitments, and management responses recalled before re-testing
External auditor correspondence — prior PBC requests, auditor findings, and response history recalled before each cycle
Entity-level controls memory — tone-at-the-top evidence and ELC assessments recalled before COSO framework reviews
IT general controls trail — ITGC test results, exceptions, and compensating controls recalled before quarterly closes
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For ISMSs managers and security compliance teams working toward or maintaining certification
ISMS scope and boundary memory — asset inventory, statement of applicability, and scope decisions recalled before audits
Risk register trail — threat assessments, risk treatment decisions, and residual risk acceptance recalled before reviews
Nonconformity and CAPA history — prior audit findings, corrective actions, and closure evidence recalled before surveillance audits
Supplier security assessment memory — third-party due diligence results and reassessment schedules recalled before renewals
Incident response log — security incidents, root cause analyses, and lessons learned recalled before tabletop exercises
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Compliance & Regulatory ·ISO 27001 & Information Security
For quality and regulatory affairs teams managing electronic records and GxP compliance
Validation history recall — IQ/OQ/PQ protocols, test results, and revalidation triggers recalled before system changes
Audit trail configuration memory — system audit trail settings and gap assessments recalled before inspections
SOP version control — current and retired SOPs, change rationale, and training dependencies recalled before updates
FDA correspondence trail — prior inspection observations, 483 responses, and commitment close-out recalled before new audits
Electronic signature governance — user authentication records and signature meaning attestations recalled before inspections
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For compliance and operations staff at electric utilities managing Critical Infrastructure Protection standards
BES Cyber Asset inventory — asset classifications, Electronic Security Perimeter boundaries, and ownership recalled before assessments
Vulnerability management history — prior patch assessments, mitigation decisions, and remediation timelines recalled before quarterly reviews
Access control and credential memory — physical and electronic access records, provisioning rationale, and revocation history recalled before audits
Incident response and reporting trail — prior reportable incidents, timelines, and lessons learned recalled before tabletop exercises
Third-party connectivity records — vendor remote access approvals, monitoring configurations, and review history recalled before renewals
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For Bank Secrecy Act officers and AML compliance teams managing transaction monitoring and regulatory obligations
SAR and CTR filing history — prior filings, disposition rationale, and follow-up activity recalled before new alerts
Customer due diligence memory — CDD and EDD decisions, risk ratings, and review history recalled before enhanced monitoring
Regulatory examination trail — prior exam findings, MRAs, and remediation commitments recalled before new examinations
Typology and red-flag library — known patterns, case precedents, and policy interpretations recalled before alert adjudication
Correspondent banking due diligence — Wolfsberg questionnaire responses and CBDDQ outcomes recalled before relationship reviews
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For ESG officers and sustainability teams managing disclosures, frameworks, and stakeholder commitments
Framework mapping memory — GRI, SASB, TCFD, and CSRD mapping decisions recalled before reporting cycles
Materiality assessment trail — prior materiality exercises, stakeholder inputs, and prioritization rationale recalled before reassessments
Emissions and data history — Scope 1/2/3 calculation methodologies, boundary decisions, and restatement rationale recalled before filings
Commitment and target tracking — public commitments, interim milestones, and progress context recalled before disclosures
Assurance engagement memory — prior assurance findings, boundary agreements, and limited vs. reasonable scopes recalled before new engagements
Memory maintenance — yours to manage in Graphnosis MemoryStudio or, by choice, deterministically with the Self-Healing Brain Engine or non-deterministically with the Neural Network (GNN) and/or Local Layer (GLL)
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For collaborative robots running pick-and-place cells with safety-rated supervision
Pre-cycle envelope check — verify the workcell perimeter, safety guards, and operator-presence sensors before each pick
Per-piece grip recall — prior grip profiles for the current part family surfaced before each pick attempt
Drift detection — vision-based pose drift compared against the baseline reference and flagged when above tolerance
Post-cycle telemetry capture — cycle time, grip success, and any protective-stop events saved for the maintenance window
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For automated vision inspection stations running on collaborative robots or fixed cells
Defect classification ground truth — prior defect signatures and accepted exemplars recalled before classifying each part
Lighting and occlusion guard — flag REVIEW when lighting drift or partial occlusion is detected, never auto-accept
Reference set recall — engineering reference geometry and tolerance bands surfaced for the active part family
Post-batch capture — new defect patterns and signature counts saved so the baseline gradually improves with every shift
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For humanoid robots handling end-to-end transport tasks in dynamic warehouse environments
Aisle clearance recall — prior clearance constraints (low headroom, fragile-stack zones, forklift overlap windows) surfaced for the destination aisle
Path planning with battery margin — plan respects active battery margin and known forklift schedules
Human-handoff protocol — request human assistance instead of replanning around blocked aisles unsupervised
Per-cycle path capture — path taken and new obstructions observed saved so the next robot in the fleet inherits the lesson
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
For mobile cobots running routine equipment-inspection walkabouts in industrial facilities
Per-asset baseline recall — baseline vibration, temperature, and acoustic signature surfaced at each stop
Drift flagging — readings more than two standard deviations from baseline flagged for the maintenance scheduler
Scope discipline — flag drift only; never attempt root-cause diagnosis or open a maintenance ticket
Post-walkabout capture — fresh reading sets saved so the baseline improves with every cycle
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Environment validator — verify all required services are healthy and the deploy window is open before any migration runs
Migration runner — apply database migrations and report each migration applied, with rollback handler on failure
Smoke test orchestrator — run the smoke-test suite against the deployed branch and surface pass/fail counts
Failure-routed rollback — On-failure handler invokes a dedicated rollback skill with the failure context preserved
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
Structured code review SOP with explicit prerequisites and a failure-recovery handler
Pre-review prerequisites — diff loaded into context and review depth stated before any analysis begins
Layered review pass — bugs, design issues, and stylistic problems surfaced as separate findings
Blocking issue counter — every reviewable change gets either an approval or an actionable comment, never silent skip
Clarification handler — pause the review and ask the user when an unclear section blocks the comment
This pack will include
–1 signed skill (.gsk pack) with SOP instructions, recall recipes, and the full 8-category goal block (Success, Out of scope, On completion, Trigger, Prerequisites, On failure, Requires, Produces)
–AI-executable via walk_skill / walk_skill_structured — any MCP-connected client can walk the skill step by step, or get a structured execution plan with declared inputs, outputs, sub-skill calls, and failure handlers
–Cross-skill orchestration ready — skills can call other skills via @skill: target(args) -> $capture syntax and route failures to a recovery skill (see the Safe Deploy pack for a worked example)
–Retrain against your own memory — the Skills Trainer personalises each skill against your cortex; snapshot-based history lets you roll back any retrain you regret
–Graphnosis-Verified (Ed25519) — every pack on this page is signed by the Graphnosis key; the App's importer verifies the signature and shows a ✓ Verified badge before any skill lands in your cortex
–GRAPHNOSIS.md companion — the pack carries a project-level GRAPHNOSIS.md you can drop into your project root, so any AI session in that project knows how to recall and save into your cortex without prior setup
–Memory maintenance — yours to manage in Graphnosis MemoryStudio; optionally delegated, deterministically, to the Self-Healing Brain Engine; or non-deterministically enriched by the Neural Network (GNN) and/or Local Layer (GLL)
The .gsk file is free and runs entirely inside the Graphnosis app — no cloud, no servers,
your memory stays yours. This kit is a starting point: every skill, engram, schema, and recall
pattern it ships with is fully expandable and customisable by you, without limitation, under the
core or any tiered plan.
Same contents as the first pack — signed .gsk skill, AI-executable, Ed25519-verified, with GRAPHNOSIS.md companion.
How Graphnosis interacts with regulated data across three operating modes.
Every mode stores data locally and encrypted on your device — the only variable
is whether an AI model also runs locally (Ollama) or via a cloud API you configure.
Graphnosis has no servers. Cloud AI is your setup, not ours.
Graphnosis never sends your data to any server. "Cloud AI" in this table refers only to cloud AI clients
you choose to configure (Claude Desktop, Cursor, ChatGPT, etc.) connected to Graphnosis via MCP.
When a cloud AI calls recall, Graphnosis returns only a small excerpt
(≤50 nodes / ≤8,000 tokens) — never the full database.
Sensitive-tier engrams send zero tokens to cloud AI by default;
a consent gate in the app fires before any sensitive data is attached to a prompt.
Regulation
Graphnosis MemoryStudio local only · no AI
Local AI + Graphnosis Ollama · air-gapped
Cloud AI + Graphnosis excerpt only · user-configured
US Healthcare & Life Sciences
HIPAA Privacy & Security Rule
✓ Compatible
✓ Compatible
Conditional¹
HITECH Act
✓ Compatible
✓ Compatible
Conditional¹
21 CFR Part 11 (FDA electronic records)
✓ Compatible
✓ Compatible
Conditional²
ICH GCP (clinical trials)
✓ Compatible
✓ Compatible
Conditional²
GINA (Genetic Information Non-discrimination Act)
✓ Compatible
✓ Compatible
Conditional¹
US Government, Defense & Intelligence
ITAR (International Traffic in Arms Regulations)
✓ Compatible
✓ Compatible
✗ Incompatible
EAR (Export Administration Regulations)
✓ Compatible
✓ Compatible
✗ Incompatible
CUI (Controlled Unclassified Information)
✓ Compatible
✓ Compatible
✗ Incompatible
DFARS / CMMC (defense contractors)
✓ Compatible
✓ Compatible
✗ Incompatible
FISMA (Federal Information Security)
✓ Compatible
✓ Compatible
✗ Incompatible⁴
SCI / Classified Programs Requires ISSO approval regardless of tool
✓ Compatible
✓ Compatible
✗ Incompatible
US Financial & Banking
GLBA (Gramm-Leach-Bliley Act)
✓ Compatible
✓ Compatible
Conditional²
SOX (Sarbanes-Oxley Act)
✓ Compatible
✓ Compatible
Conditional²
MNPI (Material Non-Public Information)
✓ Compatible
✓ Compatible
Risk³
Bank Examination / NDA-protected deal data
✓ Compatible
✓ Compatible
Risk³
US Energy & Critical Infrastructure
NERC CIP (Critical Infrastructure Protection)
✓ Compatible
✓ Compatible
✗ Incompatible
CEII (Critical Energy Infrastructure Information)
✓ Compatible
✓ Compatible
✗ Incompatible
10 CFR 50 (Nuclear Plant Safety)
✓ Compatible
✓ Compatible
✗ Incompatible
10 CFR 73 (Nuclear Security / Safeguards)
✓ Compatible
✓ Compatible
✗ Incompatible
Legal Privilege & Confidentiality
Attorney-Client Privilege
✓ Compatible
✓ Compatible
Risk³
Work Product Doctrine
✓ Compatible
✓ Compatible
Risk³
Grand Jury Secrecy (Rule 6(e))
✓ Compatible
✓ Compatible
✗ Incompatible
Tax Return Data (26 USC §6103)
✓ Compatible
✓ Compatible
✗ Incompatible
International Privacy
GDPR (EU General Data Protection Regulation)
✓ Compatible
✓ Compatible
Conditional²
UK GDPR
✓ Compatible
✓ Compatible
Conditional²
PIPEDA (Canada)
✓ Compatible
✓ Compatible
Conditional²
LGPD (Brazil)
✓ Compatible
✓ Compatible
Conditional²
PDPA (Singapore / Thailand)
✓ Compatible
✓ Compatible
Conditional²
CCPA / CPRA (California)
✓ Compatible
✓ Compatible
Conditional²
Education & Children
FERPA (Family Educational Rights and Privacy Act)
✓ Compatible
✓ Compatible
Conditional²
COPPA (Children's Online Privacy Protection)
✓ Compatible
✓ Compatible
Conditional²
Environmental & Research Data
ARPA (Archaeological Resources Protection Act) site data
✓ Compatible
✓ Compatible
✗ Incompatible
CERCLA CBI (Confidential Business Information)
✓ Compatible
✓ Compatible
Conditional²
Industry Standards & Frameworks
ISO 27001
✓ Compatible
✓ Compatible
Conditional⁴
NIST SP 800-171 / 800-53
✓ Compatible
✓ Compatible
Conditional⁴
SOC 2 Type II
✓ Compatible
✓ Compatible
Conditional⁴
Industrial Robotics & Machine Safety
ISO 10218 (Industrial robots — safety requirements)
ISO 13849 (Safety-related parts of machinery control systems)
✓ Compatible
✓ Compatible
Conditional⁴
IEC 62061 (Functional safety of machinery)
✓ Compatible
✓ Compatible
Conditional⁴
IEC 61508 (Functional safety of electrical / electronic systems)
✓ Compatible
✓ Compatible
Conditional⁴
¹ Requires a Business Associate Agreement (BAA) with your cloud AI provider. Anthropic Enterprise and OpenAI Enterprise offer BAAs.
² Requires a Data Processing Agreement (DPA) with your cloud AI provider.
³ A small recall excerpt reaches the cloud provider's infrastructure. Legal review recommended for your specific context.
⁴ Compatibility depends on your cloud AI provider's own certifications for the relevant standard.
This matrix is informational only and does not constitute legal advice. Consult your compliance officer for your specific context.
For the IT approval conversation
Business Case for Your IT Department
The questions IT will ask — and why Graphnosis passes every one of them.
🔒
Zero cloud infrastructure
Graphnosis has no servers, no sync endpoints, no ingestion API, and no telemetry. The application runs entirely on the user's device. No data leaves unless the user explicitly configures a cloud AI client — and even then, only a small recall excerpt (≤8,000 tokens) is transmitted, never the full database.
🔑
Encryption with user-controlled keys
Memory is encrypted at rest with XChaCha20-Poly1305 and an Argon2id-derived key. The key is derived from the user's passphrase on-device. It is never transmitted, never stored remotely, and never accessible to Graphnosis. IT cannot be compelled to produce keys it never had.
📋
Audit trail by design
Every write to memory is recorded in an append-only op-log with device attribution and timestamp. The log is cryptographically chained and cannot be silently modified. Exportable as a compliance artifact for HIPAA, SOX, 21 CFR Part 11, and similar frameworks.
🛡️
Sensitivity tiers and consent gates
Three tiers — public, personal, sensitive — govern how much data any AI can ever see. Sensitive-tier engrams (PHI, PII, privileged content) send zero tokens to cloud AI by default. A user-visible consent modal in the app fires before any sensitive data is attached to a prompt. IT can enforce tier policies via policy.json.
📂
Source-available — IT can audit the code
Graphnosis is licensed under FSL (Functional Source License). The complete source is public. IT security teams can inspect every line of the encryption implementation, the MCP server, the op-log, and the network layer. No black boxes, no trust-me-it's-secure.
📡
Standard open protocol — no proprietary cloud
Graphnosis exposes its 47 memory tools over MCP (Model Context Protocol), an open standard backed by Anthropic. There are no proprietary cloud endpoints, no vendor lock-in, and no hidden API calls. The MCP server runs on localhost only.
🏢
Standard desktop install
Ships as a macOS application (.dmg). No browser extensions required for core functionality. No background services that persist beyond the app lifecycle. Installs and uninstalls cleanly with no residual cloud accounts or credentials.
✈️
Fully air-gapped capable
Enable the local LLM mode (Ollama) and Graphnosis operates with no internet connection whatsoever. Every AI inference runs on-device. This is the deployment model for ITAR, CEII, classified programs, and any environment where network egress is prohibited.
Full technical documentation, architecture diagrams, and security audit guides
Individual licenses cover a single user. For firm-wide or team deployments — where multiple
people import the same schema — contact us for an organizational license.
A pre-configured starting point for Graphnosis memory in your specific domain. Each kit includes: an engram schema (which engrams to create and at which sensitivity tier); named recall patterns (the query approach for your professional context); a connector guide (how to ingest your specific sources); and an AI session template (the context file for any AI client — Claude, Cursor, Copilot, or your own agent). These kits will eventually be published as a standalone product called Job Gnosis.
Do I need a pack to use Graphnosis?
No. Graphnosis core features are free and fully functional without any pack. Packs save you the setup time of figuring out the right memory structure and recall patterns for your specific job. Everything in a pack can be built manually — the pack means you don't have to.
Why does it say zero data and no servers?
Because it's true and it matters. The pack contains no pre-loaded memory — it's a schema and a guide. Your memory is built from your own files on your own machine. Graphnosis has no cloud infrastructure: no sync server, no ingestion API, no telemetry. The source code is available to verify this.
What do I receive after purchase?
A ZIP with: the engram schema (importable directly into Graphnosis); a recall pattern guide formatted for any AI client context file; a source integration guide for the tools most common in your profession; and a setup walkthrough. Delivered immediately via Stripe.
My profession isn't listed.
Email templates@graphnosis.com with your profession and the 3–4 situations where you most need AI to remember what you know. We'll either build the pack or point you to the closest existing one.
Verify it yourself
Every claim on this page is checkable. You don't have to take our word for any of it —
the source code, the network monitor, and your own passphrase are all the proof you need.