project-jennifer
Govern Project Jennifer work through stateless-renter routing, situational CDP/CCP transitions, CEEP, POC-vs-FOC, VOC registry parsing, runtime mutation…
Skill by RobynAwesome
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/robynawesome-project-jennifer
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: ml
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
- Runs as
- Node.js package
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/project-jennifer
What's inside
- Skill name
project-jennifer- Description
- Govern Project Jennifer work through stateless-renter routing, situational CDP/CCP transitions, CEEP, POC-vs-FOC, VOC registry parsing, runtime mutation gates, NCMP, CAG, governed RAG, RIVM, NPC epistemic divergence, receipts, source authority and current repository evidence.
- License
- MIT
- Instructions
- 170 lines in SKILL.md · view on GitHub
Before installing
- Ships executable scripts
Agent Skills spec issues
- name: "project-jennifer" does not match folder "Project-Jennifer"
- allowed-tools: must be a space-separated string
Files (200)
- .env.example
- ci.yml · .github/workflows/
- deploy.yml · .github/workflows/
- jennifer-governance-python.yml · .github/workflows/
- mongodb-live-proof.yml · .github/workflows/
- mongodb-projection-rebuild-proof.yml · .github/workflows/
- postgres-api-authority-proof.yml · .github/workflows/
- postgres-live-proof.yml · .github/workflows/
- react-persisted-readthrough-proof.yml · .github/workflows/
- .gitignore
- .prettierrc.json
- 1000148049.png
- AGENTS.md
- AUGUST_2026_EXECUTION_DIRECTIVE.md
- CONTRIBUTING.md
- LICENSE
- NCMP.md
- OWNERSHIP.md
- PERN_ROADMAP.md
- Project_Jennifer.md
- README.md
- SKILL.md
- VALIDATION_FAILED.md
- VALIDATION_POLICY.md
- .env.example · apps/api/
- package.json · apps/api/
- index.ts · apps/api/src/middleware/
- mongo-projection.ts · apps/api/src/
- persistence.ts · apps/api/src/
- crisis.ts · apps/api/src/routes/
- game-continuity.ts · apps/api/src/routes/
- governance.ts · apps/api/src/routes/
- hue.ts · apps/api/src/routes/
- ingress.ts · apps/api/src/routes/
- memory.ts · apps/api/src/routes/
- ncmp.ts · apps/api/src/routes/
- relationships.ts · apps/api/src/routes/
- runtime.ts · apps/api/src/routes/
- third-signal-episode.ts · apps/api/src/routes/
- server.ts · apps/api/src/
- zero-trust.test.ts · apps/api/src/
- zero-trust.ts · apps/api/src/
- tsconfig.json · apps/api/
- .env.example · apps/web/
- .eslintrc.json · apps/web/
- jennifer-city.spec.ts · apps/web/e2e/
- next-env.d.ts · apps/web/
- next.config.js · apps/web/
- package.json · apps/web/
- playwright.config.ts · apps/web/
- postcss.config.js · apps/web/
- page.tsx · apps/web/src/app/game/consequences/
- page.tsx · apps/web/src/app/game/
- globals.css · apps/web/src/app/
- layout.tsx · apps/web/src/app/
- page.tsx · apps/web/src/app/
- page.tsx · apps/web/src/app/relationships/[relationshipId]/
- page.tsx · apps/web/src/app/relationships/
- DistrictCard.tsx · apps/web/src/components/districts/
- DistrictGrid.tsx · apps/web/src/components/districts/
- …and 140 more
SKILL.md (first lines)
# Project Jennifer
Project Jennifer is a governed intelligence runtime and portable agent-skill repository.
Use this root skill as the **public/repository entrypoint**. It intentionally gives external renters, Codex, Claude, Cursor, Gemini, Copilot and other compatible agents one stable place to discover the current Jennifer skill graph without loading the entire repository.
## Stateless-renter invariant
```text
I_AM_STATELESS_RENTER_NOT_LANDLORD
```
The human's current instruction governs task intent. `RobynAwesome/Introduction-to-MCP` supplies ecosystem doctrine when relevant. The current Project Jennifer repository supplies implementation truth. Branch, PR, CI and runtime receipts govern proof claims.
## Load order
```text
current human instruction
→ SKILL.md
→ skills.md
→ skills/project-jennifer/SKILL.md
→ skills/SKILL.md when the task enters a repository implementation lane
→ smallest relevant specialist SKILL.md
→ current implementation/source
→ source authority + privacy gates
→ execution
→ validation
→ receipt
```
Do not promote memory, semantic similarity, generated output, lore or a portable skill description into implementation truth.
## Situational conceptual routing
Project Jennifer does not grant either divergence or convergence a universal first position.
```text
CURRENT STATE
│
├─ alternatives must open/reopen ───────────► CDP / DIVERGE
├─ evidence needs evaluation ───────────────► CEEP + POC-vs-FOC
├─ stable evidence should compress ─────────► CCP / CONVERGE
└─ evidence/authority is insufficient ──────► HOLD
```
A common exploration path remains:
```text
CDP
→ CEEP
→ POC-vs-FOC
→ CCP
→ canonical/evolution receipt
```
but valid routes also include:
```text
CCP → contradictory evidence → CDP
CDP → HOLD
CCP → HOLD
```
Canonical boundary:
```text
DIVERGENCE != FOC
CONVERGENCE != POC
```
- **CDP — Conceptual Divergence Protocol:** expand or reopen the possibility space when divergence is the admitted transition. Canonical semantics + portable skill exist, and the dedicated TypeScript parser/runtime is coded and tested under `packages/conceptual/src/cdp/`; actual execution is proven only by a runtime receipt with `dedicatedCdpEngineExecuted: true`.
- **CEEP — Conceptual Evaluation Engine:** evaluate candidates and emit evaluation/evolution receipts.
- **POC-vs-FOC:** separate evidence-bearing proof from unsupported conceptual promotion.
- **VOC Registry Parser:** preserve the KPGS `POC` branch and emergent operational `FOC-G##` groups from `RobynAwesome/Introduction-to-MCP` with provenance instead of flattening them into Jennifer semantic risk categories.
- **POC/FOC Runtime Gate:** require a consequential action to survive conceptual evaluation, operational FOC checks, evidence binding and Memory Receipt admission before state mutation.
- **CCP — Conceptual Convergence Protocol:** determine which pattern currently survives evidence and evaluation; current TypeScript implementation exists under `packages/conceptual/src/ccp/`. A prior convergence may be reopened when new evidence arrives.
- **NCMP — New Concept MMAO Protocol:** govern genuinely new agent-originated concepts; human recognition remains mandatory.
- **NPC Epistemic Divergence:** `packages/npc/src/epistemic-divergence.ts` emits actor-local `CONVERGE | DIVERGE | HOLD` receipts. These are actor-model dispositions, not proof that CDP/CCP executed and not POC/FOC verdicts.
## Consequential runtime routing
Conceptual acceptance is not permission to mutate the game world.
```text
parsed VOCRegistry
→ POCFOCActionEvaluator
→ shared POCFOCActionEvaluation
→ POCFOCRuntimeGate
→ verified evidence
→ MemoryReceiptEngine
→ admitted ACCEPT: mutate once
→ HOLD / REJECT: preserve receipt, do not mutate
```
`Project Jennifer FOCType`, KPGS operational `FOC-G##`, and Memory Receipt admission are separate namespaces and must remain separately inspectable.
A player-facing consequence may be latent or initially opaque, but Project Jennifer must retain a reconstructable causal receipt for why that consequence exists.
## Specialist skill index
Start at [`skills.md`](skills.md). Major portable routes include:
- [`skills/SKILL.md`](skills/SKILL.md) — repository implementation router
- [`skills/cdp-conceptual-divergence/SKILL.md`](skills/cdp-conceptual-divergence/SKILL.md)
- [`skills/ceep-conceptual-evaluation/SKILL.md`](skills/ceep-conceptual-evaluation/SKILL.md)
- [`skills/poc-foc-registry-parser/SKILL.md`](skills/poc-foc-registry-parser/SKILL.md)
- [`skills/poc-foc-evaluation/SKILL.md`](skills/poc-foc-evaluation/SKILL.md)
- [`skills/poc-foc-runtime-gate/SKILL.md`](skills/poc-foc-runtime-gate/SKILL.md)
- [`skills/ccp-conceptual-convergence/SKILL.md`](skills/ccp-conceptual-convergence/SKILL.md)
- [`skills/ncmp-concept-intake/SKILL.md`](skills/ncmp-concept-intake/SKILL.md)
- [`skills/cag-communication-attention/SKILL.md`](skills/cag-communication-attention/SKILL.md)
- [`skills/rag-governed-retrieval/SKILL.md`](skills/rag-governed-retrieval/SKILL.md)
- [`skills/jennifer-stateless-renter/SKILL.md`](skills/jennifer-stateless-renter/SKILL.md)
- [`skills/forge-rivm/SKILL.md`](skills/forge-rivm/SKILL.md)
- [`skills/authored-relational-attention/SKILL.md`](skills/authored-relational-attention/SKILL.md)
- [`skills/jennifer-companions-npcs/SKILL.md`](skills/jennifer-companions-npcs/SKILL.md)
For the complete current registry and proof-state table, use [`skills.md`](skills.md).From RobynAwesome/Project-Jennifer, MIT.
More from RobynAwesome/Project-Jennifer
Signals
- Stars
- 7
- Forks
- 5
- Open issues
- 0
- Last commit
- Sep 14, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/RobynAwesome/Project-Jennifer
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from RobynAwesome
Cite this page
project-jennifer skill. Agent Mart. https://agentmart.dev/skills/robynawesome-project-jennifer (last updated 2026-10-24)