菜单
Run read-only OpenACCP discovery and planning. Use when facts, scope, source status, task readiness, risk, or next safe action are unclear and implementation is not yet authorized.
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。
Run delegated-final-authority OpenACCP coordination. Use for coordination decisions, authority charters, dispatch, final handoff consume, PR/CI/merge readiness, publication readiness, owner-ready waiver recommendations, and accepting or rejecting reviewed evidence when delegated by charter.
Run a bounded OpenACCP Frontier lane. Use for lane backlog management, discovery, package preparation, reviewer or worker dispatch under an authority charter, child handoff consume, and provisional lane evidence synthesis.
Perform read-only OpenACCP sidecar review of a task card, branch, diff, handoff, prompt, or status artifact. Use to check scope compliance, correctness, verification evidence, side effects, overclaiming, and readiness for final-authority consume.
Validate OpenACCP artifacts for structure, encoding, required fields, source status, authority boundary, verification evidence, overclaiming, and public-package hygiene before dispatch, handoff consume, reports, or release packaging.
Execute a narrow OpenACCP task as a scoped worker. Use when an agent must work inside assigned files, effects, workspace or branch, run focused verification, and produce a structured handoff without expanding scope.
Consume OpenACCP handoffs and decide what they prove. Use before follow-up dispatch, integration planning, final acceptance, publication, merge, amendment, or rejection.
| name | discovery-openaccp |
| description | Run read-only OpenACCP discovery and planning. Use when facts, scope, source status, task readiness, risk, or next safe action are unclear and implementation is not yet authorized. |
Read facts, classify gaps, and prepare safe next steps without implementation.
Do not edit, commit, merge, or treat reference material as current fact.
Every discovery reply must use human-explain-openaccp style in the preferred language. Explain what facts were found, what is still unproven, and which B0/B1/B2 action becomes possible next.
Every status-like discovery reply must use formal-report-openaccp structure or include a machine-summary artifact with Prompt ID, Response ID, source ids, authority, effects, basisRefs, locators, claims, and nextActions.
End every discovery reply with a short Recommended Next Step / 下一步建议 paragraph. If discovery unlocked a B0/B1/B2 action, name the agent-owned next action. If human input is needed, name the exact missing source, owner fact, repo boundary, path, approval, or decision.
Discovery output should name current, reference, deprecated, invalid, and unknown sources separately. Include locators for each material read, plus the exact next B0/B1/B2 action unlocked by those facts. Discovery does not claim implementation readiness unless a task card, authority charter, and verification plan already exist.
