Skip to content
Agent Orchestration
Skill

/failure-codifier

Extract recurring failure patterns from breezing orchestration logs and Judgment Ledger, emit failure-rule.v1 proposals with confidence scores. SSOT promotion to patterns.md or decisions.md is proposal-only — human-approval-required. Use when user mentions failure codifier,

BOOST
From plugin
claude-code-harness
3.2k23 skills5 agents5 commands
Install
$ npx -y skills add Chachamaru127/claude-code-harness --skill failure-codifier --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/failure-codifier

Context preview

The summary Claude sees to decide when to auto-load this skill.

Extract recurring failure patterns from breezing orchestration logs and Judgment Ledger, emit failure-rule.v1 proposals with confidence scores. SSOT promotion to patterns.md or decisions.md is proposal-only — human-approval-required. Use when user mentions failure codifier,

SKILL.md

failure-codifier.SKILL.md
name: failure-codifier
description: "Extract recurring failure patterns from breezing orchestration logs and Judgment Ledger, emit failure-rule.v1 proposals with confidence scores. SSOT promotion to patterns.md or decisions.md is proposal-only — human-approval-required. Use when user mentions failure codifier, failure patterns, self-learning loop, codify failures, or failure-rule proposals. Do NOT load for: direct SSOT edits, auto-promotion, or implementation unrelated to failure analysis."
description-en: "Extract recurring failure patterns from breezing orchestration logs and Judgment Ledger, emit failure-rule.v1 proposals with confidence scores. SSOT promotion to patterns.md or decisions.md is proposal-only — human-approval-required. Use when user mentions failure codifier, failure patterns, self-learning loop, codify failures, or failure-rule proposals. Do NOT load for: direct SSOT edits, auto-promotion, or implementation unrelated to failure analysis."
description-ja: "breezing ログと Judgment Ledger から再現失敗パターンを抽出し、confidence 付き failure-rule.v1 候補を提案する。patterns.md / decisions.md への昇格は human-approval-required(自動昇格禁止)。Use when user mentions failure codifier, 失敗パターン, self-learning loop, codify failures. Do NOT load for: SSOT 直接編集, 自動昇格."
allowed-tools: ["Read", "Bash", "Grep"]
argument-hint: "[propose|explain]"
user-invocable: true

Failure Codifier

breezing orchestration ledger + Judgment Ledger から再現失敗を **read-only** で抽出し、`failure-rule.v1` 候補を confidence score 付きで提案します。

核心契約

  • **human-approval-required**: codifier は dry-run 提案のみ。`patterns.md` / `decisions.md` への自動昇格は構造的に禁止。
  • Confidence 閾値: occurrence **count ≥ 3 → medium**、**count ≥ 5 → high**(`go/internal/failurecodifier/confidence.go`)。
  • 昇格先 heuristic: `proposed_ssot_target` フィールドで `patterns.md` または `decisions.md` を**提案するのみ**。

使い方

Dry-run 提案(推奨)

./scripts/failure-codifier-propose.sh --dry-run

stdout は JSON 配列(`failure-rule.v1` 候補)。SSOT ファイルには一切書き込みません。

Go テスト

cd go && go test ./internal/failurecodifier/... -count=1

参照

  • 昇格ワークフロー: [references/promotion-workflow.md](${CLAUDE_SKILL_DIR}/references/promotion-workflow.md)
  • Schema: `templates/schemas/failure-rule.v1.json`
  • Core: `go/internal/failurecodifier/`

禁止事項

  • `patterns.md` / `decisions.md` への Write / Edit(human 承認後も codifier 経由では不可)
  • `AutoPromote` / unattended SSOT 更新
  • Plans.md の `cc:*` マーカー変更
Read more
Ships withclaude-code-harness

Plan. Work. Review. Ship. A disciplined delivery loop for Claude Code, Codex CLI, Cursor, and Grok.

Get the whole plugin

Other skills on claude-code-harness.