Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add skills/morodomi/dev-crew/codify-insightnpx skills add morodomi/dev-crew --skill codify-insightgit clone --depth 1 https://github.com/morodomi/dev-crewWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00089 | $0.01183 |
| Opus 5 | $0.00044 | $0.00592 |
| Sonnet 5 | $0.00018 | $0.00237 |
| Haiku 4.5 | $0.00009 | $0.00118 |
Grade A, and why
codify-insight scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured yesterday.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 98 lines — stays where its author put it; the contents beside it link to each section on GitHub.
codify-insight
Workflow
Hard Gate
Captured cycles scan (frontmatter-only):
for f in docs/cycles/*.md; do
awk '/^---$/{c++;next} c==1{print}' "$f" | grep -q 'retro_status: captured' && echo "$f"
done
非空 → 処理対象 cycles を収集。 空 → no-op exit 0: 「No captured cycles found. Nothing to codify.」を出力して終了。
Idempotency Check
各 cycle doc の retro_status を確認:
retro_status: resolved→ skip(既に全 insight 判定済み)retro_status: none→ skip(retrospective 未実行)retro_status: captured→ 処理対象
## Codify Decisions セクション内の ### Insight N heading 存在 = 該当 insight 判定済み。
partial 完了時は未判定 insight のみ処理し、captured を維持して再起動で残分を処理。
Recurrence-aware Pre-triage (質問を減らす一次フィルタ)
autonomous triage の前に、直近 10 cycle docs の ## Retrospective + ## Codify Decisions を scan し頻度判定:
- 同 key phrase が 2+ 回再発 → 自動
codified→rule(質問スキップ、promotion 確定) - 過去 cycle で同 insight が
no-codify判定済み → 自動no-codify(質問スキップ、duplicate) - 1 回以下 (novel / 初出) → 下記 autonomous triage へ
scan 詳細 + threshold 調整は reference.md 参照。
Default: Autonomous Batch Triage
pre-triage で処理されなかった insight は 1 cycle 分まとめて自動 triage。per-insight AskUserQuestion をデフォルトにしない。
優先ルール:
- 次 cycle の TDD をすぐ harden できるもの →
codifiedwithruleorinline-update - post-TDD の広い改善、repo-wide sweep →
deferred(通常new-cycle) - observation-only / 2nd-order note →
no-codify skilldestination は rare。2+ cycles 再発 + 複数 phase / repo 再利用時のみ- destination
ruleは tier(always/cycle-scoped/file-scoped)を必須指定する - tier ごとの paths・交換条件・反映責務は reference.md に従う
AskUserQuestion Fallback (only when needed)
以下の場合のみ AskUserQuestion (1 cycle max 1 回の batch、per-insight 不可):
skill候補で destination 妥当性の確認が必要- autonomous triage で
codifiedvsdeferredの境界が低確信 - ユーザーが対話的 codify を明示要求
全件 recurrence or high-confidence triage なら質問 0 件で summary のみ print。
Output
Cycle doc EOF に ## Codify Decisions セクションを append (APPEND-ONLY):
## Codify Decisions
### Insight N
- **Decision**: codified
- **Destination**: rule
- **Tier**: cycle-scoped (rule のみ必須: always / cycle-scoped / file-scoped)
- **Paths**: (file-scoped のみ必須: repo-relative glob)
- **Reason**: (optional)
- **Decided**: YYYY-MM-DD HH:MM
What ships with it
1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- yesterday First seen · 98 lines · 89 tokens per session scan A 69df2081caff
codify-insight is a skill published in the GitHub repository morodomi/dev-crew (1 stars, last pushed 3d ago), licensed MIT. It adds 89 tokens to every session and 1,183 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
nw-collapse-detection
Documentation collapse anti-patterns - detection rules, bad examples, and remediation strategies for type-mixing violations.
sw-learn
Captures patterns and learnings from the current work unit. Reviews build failures, gate findings, and architecture decisions. Applies objective promotion criteria autonomously — patterns.md is the artifact.
article-writing
Write articles, guides, blog posts, tutorials, newsletter issues, and other long-form content in a distinctive voice derived from supplied examples or brand guidance. Use when the user wants polished written content longer than a paragraph, especially when voice consistency, structure, and credibility matter.
a-evolve
Apply A-Evolve's agentic evolution methodology to improve AI agent performance across runs. Use when the user wants to diagnose agent failures, generate targeted skills from error patterns, evolve system prompts, or accumulate episodic knowledge. Works standalone or inside AutoResearchClaw pipelines. Triggers on…
hive.chart-creation-foundations
Required reading whenever any chart tool is available. Teaches the one-tool embedding contract (call chartrender → live chart appears in chat AND a downloadable PNG lands in the queen session dir), the ECharts (data viz) vs Mermaid (structural diagrams) decision, the BI/financial-grade aesthetic baseline (no…
📝 任务完成后归档
重要提醒: 每次完成复杂调试或开发任务后,主动执行此流程! 将学到的经验归档为 skill,供以后参考。不要等用户提醒。.