Menu

Individual skill

codexqa-code-reviewer

Graph-evidence AI code review (codexqa-code-reviewer) for ANY language repo using ONLY the CodexQA CLI symbol graph (call chains, classes, methods, configs, blast radius, test edges), then an order-16 Agent LLM judgment pass by the host agent's embedded model with deterministic dedupe/merge against heuristic findings. Use when the user asks for codexqa-code-reviewer (former name ai-code-reviewer), code review, PR review, 代码评审, impact analysis, 影响面, regression scope, test gaps, full-repo health review, 全仓评审, CodexQA evidence-pack, 证据包, graph-backed review, or LLM semantic CR on a pack. Requires codexqa CLI for every language; never embeds CodexQA source; never substitutes git-diff-only analysis. Not SAST+agent scan reports (that is codexqa-defect-analyzer), not structure/impact mapping alone (that is codexqa-code-analyzer), and not exception RCA (that is codexqa-rootcause-analyzer).

skills/codexqa-code-reviewer/SKILL.md

Read pinned source

Source commit 783954281fc4898a9fac66a575363f419d4683e4

What was checked

Matches the observed repository HEAD
Yes
Scanner rules are current
Yes
Repository scan complete
No
Skill scan complete
No
Observed behavior
Network or broad shell
Last checked
Nov 5, 2026

These are observations at the recorded commit, not a guarantee of safety. Review the source before running it.

Install this skill

Choose an agent in your Stack or copy a command below. Commands fetch the pinned commit and copy only this skill directory. Existing destinations are refused. Copying a command does not install anything.

An install command is unavailable: current rules, complete repository and skill scans, an eligible behavior level and a permitted license are required.

Repository context

Repository stars (not skill ratings)
236
Repository last updated
Nov 6, 2026
License
Apache-2.0

Observed agent integration: No specific integration evidence. Standard SKILL.md installation does not imply verified compatibility.

Files and excerpt

File metadata and permitted, redacted excerpts come from the recorded scan. Source links are pinned to that commit.

Browse files
Read excerpt

# AI Code Reviewer

Graph-first code review via **CodexQA CLI** only. Collect a JSON evidence pack,
then reason from those artifacts. Applies to every language / polyglot monorepo.

CLI: `{baseDir}/scripts/collect-pr-evidence.sh` (and full-repo / adhoc variants).
Runtime pack: `<repo>/.codexqa-review/<run-id>/` → `review-conclusion.json` +
`REVIEW-REPORT.html`.

**Install:** Prefer `npx skills add openqa-cn/codexqa --skill codexqa-code-reviewer`.
Do not copy into a skills library manually until the user names the install target.

`README.md` / `README.zh-CN.md` / `HOW_IT_WORKS.md` / `KNOWN_LIMITATIONS.md`
(and their `.zh-CN` twins) are human-facing. Do not load them at runtime.