원클릭으로
demerzel
Talk to Demerzel — the AI governance agent. Query beliefs, run pipelines, execute epistemic commands, and get recommendations via ACP.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Talk to Demerzel — the AI governance agent. Query beliefs, run pipelines, execute epistemic commands, and get recommendations via ACP.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | demerzel |
| description | Talk to Demerzel — the AI governance agent. Query beliefs, run pipelines, execute epistemic commands, and get recommendations via ACP. |
| allowed-tools | Read, Write, Edit, Bash, Grep, Glob |
Talk to Demerzel via the ACP (Agent Communication Protocol) bridge. Demerzel runs as a standalone agent on port 8200 with 4 sub-agents.
cd Apps/demerzel-agent && DEMERZEL_API_KEY=mykey uvicorn src.server:app --port 8200
dotnet build Apps/demerzel-bridge/DemerzelBridge/DemerzelBridge.csproj
.mcp.json (already done — "demerzel" entry)Parse the user's input and route to the correct sub-command:
| User says | Route to | MCP tool |
|---|---|---|
/demerzel beliefs or "list beliefs" | Governance | demerzel_governance |
/demerzel policies | Governance | demerzel_governance |
/demerzel constitution [name] | Governance | demerzel_governance |
/demerzel strategies | Governance | demerzel_governance |
/demerzel brainstorm <title> | Pipeline | demerzel_pipeline |
/demerzel plan <title> | Pipeline | demerzel_pipeline |
/demerzel build <title> | Pipeline | demerzel_pipeline |
/demerzel review <title> | Pipeline | demerzel_pipeline |
/demerzel compound <title> | Pipeline | demerzel_pipeline |
/demerzel run all <title> | Pipeline (full) | demerzel_pipeline |
/demerzel show tensor | Epistemic | demerzel_epistemic |
/demerzel show beliefs where ... | Epistemic | demerzel_epistemic |
/demerzel methylate <id> [reason] | Epistemic | demerzel_epistemic |
/demerzel demethylate <id> | Epistemic | demerzel_epistemic |
/demerzel amnesia <belief> [in N days] | Epistemic | demerzel_epistemic |
/demerzel broadcast | Epistemic | demerzel_epistemic |
/demerzel whats next or /demerzel (no args) | What's Next | demerzel_whats_next |
| Any other text | What's Next (with query) | demerzel_whats_next |
curl -sf http://localhost:8200/agents | head -1 && echo "OK" || echo "NOT RUNNING"
If not running, tell the user:
Demerzel ACP agent is not running. Start it with:
cd Apps/demerzel-agent && uvicorn src.server:app --port 8200
If the MCP tool demerzel_governance / demerzel_pipeline / demerzel_epistemic / demerzel_whats_next is available, call it directly.
If MCP tools are not available (bridge not running), fall back to direct HTTP:
curl -s -X POST http://localhost:8200/runs \
-H "Content-Type: application/json" \
-H "Authorization: Bearer ${DEMERZEL_API_KEY}" \
-d '{"agent_name": "<agent>", "input": [{"parts": [{"content": "<message>", "content_type": "text/plain"}]}]}'
After each command, suggest related actions:
list beliefs → "Try show tensor to see the epistemic distribution"whats next → "Pick an item and run brainstorm: <title> to start"brainstorm → "Run plan: <title> to create an implementation plan"compound: <title> to document learnings"/demerzel skill
│
├── MCP path: Claude Code → DemerzelBridge (MCP/stdio) → ACP HTTP → Demerzel Agent
│
└── HTTP fallback: Claude Code → curl → ACP HTTP → Demerzel Agent
The epistemic commands implement Articles E-0 through E-9:
| Article | Command |
|---|---|
| E-1 Contradictory Ground | show beliefs where truthState = contradictory |
| E-3 Epistemic Viscosity | show beliefs where viscosity > 0.8 |
| E-5 Deliberate Amnesia | amnesia <belief> in 30 days |
| E-7 Epistemic Tensor | show tensor |
| E-8 Epigenetics | methylate <strategy> reason ... |
| E-9 Federated Review | broadcast beliefs |
/demerzel
→ Scans GitHub + CI/CD + governance, returns prioritized recommendations
/demerzel beliefs
→ Lists all tetravalent belief states (T/F/U/C)
/demerzel show tensor
→ Shows epistemic tensor distribution (wisdom, hunches, blindspots)
/demerzel brainstorm: Godot bridge protocol
→ Runs brainstorm stage via Ollama, returns 3-5 ideas
/demerzel run all: Fix CI failures
→ Runs full pipeline: brainstorm → plan → build → review → compound
/demerzel methylate socratic_pathogen reason caused learner distress
→ Suppresses the strategy (Article E-8)
/demerzel amnesia old-architecture-assumption in 14 days
→ Schedules belief for deletion test (Article E-5)
Post-merge intent-vs-delivery evaluator. Resolves the most recent squash-merged PR, captures its stated intent (title + Summary), diffs it against what actually landed, runs /octo:review for specialist commentary, scans Codex bot comments (P0/P1 degrade the grade), computes a high/medium/low alignment score, and writes a structured grade to state/quality/pr-grades/<merge-sha>.json. Closes the "agent declared done, nobody checked the goal" loop.
Run one bounded supervised autonomous cycle on the Guitar Alchemist repo. Picks the smallest unchecked roadmap slice inside allow_edit, implements it, runs the oracle, emits cycle evidence, and stops. Refuses to run unless dev-process-overseer reports loop-eligible and the preflight passes. Use when asked to "run one supervised loop", "advance the roadmap autonomously", or "do one cycle and stop".
Rebuild the OPTIC-K voicing index (state/voicings/optick.index). Invoke when schema hash bumps, tag enrichment changes, new chord qualities are added, or the leak-test / invariant-coverage tools flag regressions. Mandatory precondition: stop any process that has the index mmap-locked — GaApi and GaMcpServer both hold the file open at runtime, and the write step WILL fail with IOException otherwise.
Generalized Cherny-style autonomous improvement loop. Reads a domain-scoped baseline.json + oracle script, runs the oracle, picks the worst-scoring item, proposes a fix, validates rollback safety, commits if metric improved AND roundtrip passes. Use for chatbot-qa, embeddings, voicing-analysis, or any other domain that has a baseline + oracle. Refuses to run if killswitch present, governance halt active, or scope_boundary violated.
End-to-end autonomous (AFK) harness that develops AND QAs the GA chatbot. Orchestrates the existing Level-3 /auto-optimize dev loop + the semantic /ga-chatbot-qa-panel judge workflow, instruments every step to a dashboard the human can watch, and enforces branch-only / never-merge safety. Use to run unattended chatbot-quality improvement with full human visibility. Refuses to run if the backend is down, a killswitch/halt marker is present, or scope_boundary is violated.
Turns raw Sentrux structural-quality metrics (quality_signal, cycle count, coverage) into a ranked list of actionable refactor recommendations with starter sketches. Pulls live data from sentrux MCP tools, optionally cross-references AI annotations (@ai:business-value, @ai:smell) and the latest /test-plan output, then writes 5-10 prescriptive bullets to state/quality/sentrux-next-steps/<date-Z>.md so the Sentrux dashboard tab can render them. PROPOSES — never auto-refactors; the human picks which recommendation to take.