ワンクリックで
wrought-implement
Start an autonomous implementation loop with test verification
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
メニュー
Start an autonomous implementation loop with test verification
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
Strategic system analysis for health assessment, pattern recognition, component deep dives, risk assessment, and architecture review. Use for proactive system understanding beyond reactive incident response.
Transform a design into an implementation spec. Creates a blueprint document and an implementation prompt for /plan mode. Use after /design or /research.
Cancel the active Wrought implementation/RCA loop
Interactive design analysis for architecture decisions. Combines codebase analysis, documentation review, and external research to provide evidence-based recommendations. Use for tradeoff analysis, pattern evaluation, migration planning, or design validation.
Log a proactive discovery that needs attention. Creates a factual record of WHAT was found with classification and a Findings Tracker. Use for analysis discoveries, review findings, or inspection results.
Deep multi-agent code review with algorithmic complexity analysis, data structure review, paradigm enforcement, and efficiency analysis. Complements /simplify with deeper dimensions.
SOC 職業分類に基づく
| name | wrought-implement |
| description | Start an autonomous implementation loop with test verification |
| disable-model-invocation | false |
| wrought | {"version":"1.0","platforms":{"claude-code":{"disable-model-invocation":false}}} |
You are starting a Ralph Wiggum inner loop for implementation. A Stop hook will automatically verify your work after each attempt to stop.
Read .claude/wrought-loop-state.json. If it exists and active is true:
/cancel-wrought-loop first, or continue the existing loop."The user should provide a plan path. If not provided:
docs/plans/*.mdRead the .wrought marker file and extract:
verifier_type — if none, ask the user: "What verifier command should I use? (e.g., uv run pytest tests/ -q)"verifier_test_path — default test pathverifier_max_loops — iteration budgetverifier_timeout — per-run timeout in secondsWrite .claude/wrought-loop-state.json:
{
"active": true,
"preset": "wrought-implement",
"finding_id": "{from plan or Findings Tracker}",
"tracker_path": "{path to active FINDINGS_TRACKER.md}",
"verifier_command": "{verifier command}",
"max_iterations": {from marker},
"current_iteration": 0,
"started_at": "{ISO timestamp}",
"history": []
}
When the loop completes (verifier passes):
.claude/wrought-loop-state.json to summarize iterationsdocs/capsules/{finding_id}/"/forge-review --scope=diff to verify code quality."