القائمة
Read-only code review of jj changes.
التثبيت باستخدام Codex أو Claude انسخ هذا Prompt والصقه في Codex أو Claude أو مساعد آخر ليراجع صفحة Skill ويثبّتها لك.
استنادا إلى تصنيف SOC المهني
| name | review |
| description | Read-only code review of jj changes. |
Analyze jj changes and produce categorized findings. No tree manipulation, no side effects.
Do NOT activate for trivial changes (< 5 files, < 50 lines) — just review inline.
jj diff -r 'bookmark' # bookmark
jj diff -r 'change-id' # change ID
jj diff --from 'A' --to 'B' # range
jj log -r 'target' -T 'description' --no-graph
jj diff --stat
jj diff
Read full files with read tool — do NOT review from diffs alone.
Classify each finding:
| Category | Emoji | Meaning |
|---|---|---|
| Must fix | 🔴 | Bug, security issue, broken logic |
| Suggestion | 🟡 | Better approach, missing edge case |
| Nit | 🟢 | Style, naming, minor cleanup |
For each: file/line range, what, why (one sentence), concrete fix.
Verdict: Request Changes (any 🔴) | Comment (only 🟡) | Approve (only 🟢 or none)
## Review: <change description>
**Verdict:** Request Changes / Comment / Approve
**Summary:** <1–2 sentence overview>
---
### 🔴 Must Fix
**`file.ts:42–55`** — <what>
<why it matters>
```suggestion
<concrete fix>
file.ts:88 —
file.ts:12 —
Files reviewed: /
## Usage
/skill:review # current change /skill:review main # bookmark /skill:review kowqznzo # change ID /skill:review abc::def # range
Browser automation via agent-browser. Use when the user wants to browse, interact with, or inspect web pages.
Review agent-facing knowledge (rules, AGENTS.md, CONVENTIONS.md, README.md, domain docs) for staleness and accuracy. Use when the user asks to audit, clean, or prune knowledge. Also use when they're confused by a rule/doc that seems wrong — treat that as the starting point for a broader audit.
Reflect on session, update rules, commit with jj.
Adversarial design interview.
Ingest sources into the wiki.
Wiki health checks.