メニュー
Extract non-obvious learnings from sessions into structured knowledge artifacts
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。
SOC 職業分類に基づく
Use when you have a written implementation plan to execute
Spreadsheet toolkit. Reach for it whenever the artifact on either side of the conversation is a workbook file — .xlsx, .xlsm, .xltx, .csv, .tsv — and the user wants that artifact produced, changed, cleaned, or read. Typical triggers: 'build me a model', 'update this sheet', 'add a column', 'compute the totals as formulas', 'sanity-check this xlsx', 'export sheet 2 to CSV', 'render the workbook as PDF', 'the spreadsheet in ~/Downloads is a mess, fix it'. Applies equally to financial models, ops reports, data cleanups, and template fills. Skip it when the workbook is only source material and the real output is a Word doc, an HTML page, a Python script that runs standalone, a Google Sheets integration, or an ingestion pipeline into a database — in those cases the spreadsheet is a means, not the deliverable.
Perform a focused, evidence-based code review. Use when the user asks to review code, a diff, pull request, branch, commit, or implementation before merging.
Smart git commit message generator following Conventional Commits. Use when the user says 'commit', 'commit this', 'write a commit message', 'smart commit', or asks to create a git commit.
Use when the user asks what OpenZeroCode can do, how a feature works (memory, checkpoints, agents, subagents, tasks, compose, learn), how to configure it, where config/data lives, which config key controls a behavior, what CLI or slash commands exist, or how to enable/disable/tune something — the self-documenting reference for OpenZeroCode itself.
Use when executing implementation plans with independent tasks in the current session
| name | compose:learn |
| description | Extract non-obvious learnings from sessions into structured knowledge artifacts |
Extract durable, non-obvious knowledge from completed work and persist it as structured artifacts.
Core principle: Only extract what a future agent wouldn't know from reading the code or docs.
Include (non-obvious discoveries only):
Exclude:
Review the session for:
| Scope | Target | When |
|---|---|---|
| Project | docs/compose/learnings/PROJECT.md | Codebase-specific patterns |
| Global | ~/.openzerocode/LEARNINGS.md | Cross-project patterns |
## [YYYY-MM-DD] <short description>
**Scope:** project | global
**Confidence:** low | medium | high
### Observation
<What was discovered>
### Evidence
<How you know this>
### Implication
<What this means for future work>
| Level | Meaning |
|---|---|
| low | First observation, single occurrence |
| medium | Multiple occurrences, clear evidence |
| high | Consistently observed, well-understood |