菜单
Portable orchestration skill for pi. Use when Claude and Codex should work as peer agents with bounded scopes, stable task cards, and evidence-backed handoffs. Use it to drive the orchestrator -> builder -> reviewer loop and keep all delegated work tied to plans/<feature>/ artifacts.
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
Portable builder skill for pi. Use when a bounded task card is approved and a builder must execute one artifact-first loop at a time against plans/<feature>/ artifacts under orchestrator and reviewer oversight.
Local Effect doctrine for the pi healthcare harness. Use when a task touches files importing effect and needs consistent rules for services, schemas, errors, layers, config, and structured effectful composition.
Practical FPF usage guide for the vfpf harness. Use when a task needs a fit check, an FPF entry route, starter artifacts, mixed-team handoff rules, or an FPF-shaped review rubric without loading the full specification into the prompt.
Portable evaluation skill for pi. Use when comparing runs, agent mixes, or orchestration changes for Claude and Codex. Only compare results inside an explicit comparability frame and tie outputs back to plans/<feature>/.
Portable planning skill for pi. Use when a feature needs bounded context, explicit alternatives, and one approved execution plan before implementation. Keep the output artifact-first so Claude and Codex can work against the same plan surface. Write plans under plans/<feature>/.
Portable review skill for pi. Use when work needs a findings-first review that checks behavior, regression risk, evidence quality, and plan adherence across Claude and Codex outputs. Review against plans/<feature>/ artifacts and act as the blocking reviewer in the orchestrator loop.
| name | vfpf-orchestrate |
| version | 0.1.0 |
| description | Portable orchestration skill for pi. Use when Claude and Codex should work as peer agents with bounded scopes, stable task cards, and evidence-backed handoffs. Use it to drive the orchestrator -> builder -> reviewer loop and keep all delegated work tied to plans/<feature>/ artifacts. |
| allowed-tools | ["Read","Write","Edit"] |
Coordinate multi-agent work without losing scope, evidence, or ownership.
plans/<feature>/tasks/<task-id>.md for task cards and handoffs.effect-best-practices
plus plans/<feature>/effect-review-rubric.md for those loops.plans/<feature>/sprint-contract.md to lock the current red-green loop.plans/<feature>/review-report.md and
plans/<feature>/findings-absorption.md to track blocked review cycles.plans/<feature>/external-review-bundle.md before any human or
external review boundary.plans/<feature>/plan.mdplans/<feature>/tasks/<task-id>.mdplans/<feature>/sprint-contract.mdplans/<feature>/review-report.mdplans/<feature>/findings-absorption.mdplans/<feature>/decisions.mdplans/<feature>/verification.md