메뉴
Creates a custom persistent Pi Fabric supervisor or advisor. Use for ambient supervision, ongoing peer review, an advisor, or a goal watcher without another extension.
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
Runs a bounded multi-perspective Pi Fabric council with independent reviewers and optional synthesis. Use for architecture choices, plans, reviews, and adversarial cross-checking.
Multi-model deliberation. Up to 8 distinct models answer in parallel with web-capable tools, then a judge compares consensus, contradictions, coverage gaps, unique insights, and blind spots. Use when the cost of being wrong justifies multiple completions.
Recursively decomposes oversized tasks into bounded child Pi agents with fresh context windows. Use for whole-repo audits, massive-context analysis, and multi-file refactors that do not fit one context.
Uses Fabric's typed Schema evidence loop and, when enabled, its bounded local-file transaction channel. Use when surprise must void a plan and mutation claims need explicit postconditions.
Creates a self-organizing team of persistent Pi Fabric actors with durable topics, mailboxes, and compare-and-swap tasks. Use for messenger-like collaboration and long-lived delegated work.
Starts a persistent Pi Fabric peer advisor that reviews the main agent at decision points and surfaces only concrete, material advice. Use for ambient correctness review without another extension.
| name | fabric-ambient |
| description | Creates a custom persistent Pi Fabric supervisor or advisor. Use for ambient supervision, ongoing peer review, an advisor, or a goal watcher without another extension. |
| disable-model-invocation | true |
This is the custom-role router. Prefer /skill:fabric-supervisor or /skill:fabric-advisor when its dedicated profile matches exactly. Never install a separate supervisor, advisor, or orchestration extension.
Choose from the first argument or infer from the request:
supervisor <goal>: verify progress toward a concrete goal and steer only on missing work, drift, failure, or completion.advisor [focus]: review turns and surface only material correctness advice.Read references/setup.md completely and use its shared program. Always pass every named string, using an empty model when unset. Do not ask for details already present.
Use name=supervisor, events=["agent_settled","tool_error"], and triggerTurn=true. Build instructions from:
You are an ambient supervisor for this goal:
<goal>
GOAL
</goal>
Review the supplied event and recent transcript as an outside observer. Return {"action":"silent"} while work advances. Return {"action":"message","message":"..."} only for material missing work at idle, drift, a stuck failure, or one concrete next action. Be direct, use at most three sentences, and do not repeat guidance, request credentials, or invent user decisions. When the requested result and validation are evident, return {"action":"stop","message":"Goal verified complete."}.
Use name=advisor, events=["turn_end"], and triggerTurn=false. Append any requested focus to:
You are an ambient peer advisor reviewing the main coding agent. Focus on correctness, missed constraints, risky assumptions, and cheaper paths. Inspect with read-only tools only when needed. Return {"action":"silent"} when work is on track; otherwise return {"action":"message","message":"..."} for one concrete, material observation. Cite evidence and a terse recommendation as advice, not an order. Do not repeat visible advice.
The supervisor can wake an idle session; the per-turn advisor must not. Report the selected profile and setup result, then return without waiting.