本文へ移動
cccskills
無料GitHub で公開

smoke-green-loop

Diagnose and repair a failed GEODE smoke workflow within the requested acceptance criteria. Preserve failed evidence, verify the shared root cause, and apply merge, rebuild, or live rerun stages only when authorized.

インストール方法を見る

含まれるファイル(1)

  • SKILL.md1.6 KB

SKILL.md(原文)

インストールする前に、エージェントに与えられる指示の中身を確認できます。

Smoke-Green Loop

Use .agents/skills/geode-workflow/SKILL.md and the repository GitFlow. For each iteration:

  1. Preserve the failed run and read its structured state, transcript, per-agent results, and diagnostics. Never delete evidence to make a rerun green.
  2. Identify one root cause and the smallest reproduction. Distinguish runtime failure from quota, credentials, remote service, or harness contamination.
  3. Fix the shared root, add one regression check, and run targeted static and test gates. Update CHANGELOG.md for functional changes.
  4. If integration is authorized, follow the repository GitFlow and required CI. Rebuild or restart only when in scope; a local fix can be verified in its owned worktree without first merging or changing a global installation.
  5. Archive the prior run, obtain explicit approval for any live or paid call, and rerun with the same frozen inputs unless the hypothesis explicitly changes them.

Declare success only when the specified acceptance checks pass and requested artifacts contain the expected evidence. Preserve historical failure events; they do not need to disappear for a later attempt to succeed. Stop and report when remaining work needs new authority, user input, or an external-state change. Report attempted iterations, invalid attempts, costs, and unresolved limits; do not select only the first green run.

レビュー

まだレビューはありません。使ってみた感想をお寄せください。

同じリポジトリのスキル

概要と使いどころ

Audit agent runtimes and their tests for lifecycle inference, capability overload, context pollution, correlated verification, prompt-only control, false success, dead compatibility paths, test-only seams, and complexity without contribution. Use for whole-repository agent audits, slop or dead-code triage, test contribution reviews, large refactors, tool/context simplification, and cleanup requests that need proof before deleting code or tests.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

Diagnose agent operational failures through default-policy checks, multi-gap analysis, explicit service ownership, request-context lifetime, and degradation-versus-correctness checks.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

Align, run, audit, and publish GEODE benchmark comparisons against Agent-World arXiv 2604.18292v1 across MCP-Mark, BFCL V4, and tau2-bench. Use for Agent-World or AgentWorld comparison requests, cross-suite agent benchmark design, Tau2 comparator/profile diagnosis, paired thin-runtime-versus-GEODE experiments, mean-accuracy@8 replication, or benchmark comparability and artifact reviews.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

Verification checklist to prevent fake success. Detects test deletion/disabling, coverage regression, lint bypass, secret exposure. Triggered by "deception", "fake" ("가짜"), "fake success", "verification" ("검증"), "checklist" ("체크리스트"), "deletion detection" ("삭제 탐지"), "regression" keywords.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

Implement proven backend architecture patterns including Clean Architecture, Hexagonal Architecture, and Domain-Driven Design. Use when architecting complex backend systems or refactoring existing applications for better maintainability.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

Auto-search and summarize latest AI/agent papers.

日本語の概要は準備中です。原文の説明を表示しています。

mangowhoiscloud/geode152026年10月12日 更新

mangowhoiscloud のスキルをすべて見る

このスキルの問題を報告する