codexqa-defect-analyzer skillA
codexqa-defect-analyzer is agent-read markdown (skill) from openqa-cn/codexqa: AI defect detection and code-risk scanning for engineering projects. Use when the user asks for AI defect detection, code-risk scan, review a diff/MR/PR/commit (incremental), baseline-scan a repository (full), OR upload/paste code snippets for adhoc defect detection. Combines two detection dimensions — (1) deterministic SAST/lint/secrets/SCA and (2) one-round Agent LLM Detection by the invoking agent's embedded model (no separate LLM API key) — then dedupes and merges them into the final report..
Indexed from public GitHub and served as immutable, content-addressed versions. Install it pinned to an exact SHA-256 with the mdr CLI, and every file is verified against the hash recorded here before it reaches your agent. The deterministic audit below grades the latest version, and the same file always earns the same grade.
What the file says
# Defect Detection
You are a codexqa-defect-analyzer orchestrator. Commands are English; user-facing text may be Chinese.
**Never** start semantic review before deterministic collect + context budget checks.
CLI: `python3 {baseDir}/scripts/run_scan.py` (`$SKILL_SCRIPT`). Runtime artifacts: `-o` report dir (default `/tmp/aid_report/`) and optional `{baseDir}/data/` feedback.
`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.
## Boundaries
| Need | Skill |
|---|---|
| SAST + Agent LLM Detection → `report_scan.*` (P0–P3) | **this skill** (`codexqa-defect-analyzer`) |
| CodexQA evidence-pack + Agent LLM judgment → bilingual `REVIEW-REPORT.html` | `codexqa-code-reviewer` |
| Symbol-graph change impact, callers, test gaps | `codexqa-code-analyzer` |
| Exception RCA from stacks/logs on top of CLI analysis | `codexqa-rootcause-analyzer` |
## Detection dimensions
Final `report_scan.*` = **deduped merge** of two dimensions:
| Dimension | Who runs it | Output / provenance |
|-----------|-------------|---------------------|
…Read the whole file at its exact version.
How to install
mdr add openqa-cn/codexqa/codexqa-defect-analyzer@v0.0.2mdr add openqa-cn/codexqa/codexqa-defect-analyzer@sha256:c4395343b67d92e1Pin to a label to follow the author's releases, or to a sha256 to freeze the exact bytes forever. Either way the resolved hash is written to mdr.lock, and mdr install reproduces it on any machine.
[](https://markdownregistry.com/a/art_aupfxdbcidry27wx)
1 badge views in 30 days
Versions
Audit of the latest version
- pass: Frontmatter block present
- pass: Frontmatter declares a name
- pass: Frontmatter declares a description
- pass: Size between 200 bytes and 200 KB (14444 bytes)
- pass: No zero-width or bidi control characters
- pass: No instruction hidden inside an HTML comment
- pass: No link to an exfiltration or paste host
- pass: No credential-shaped string
- pass: No instruction to send local credentials anywhere
- pass: No text hidden with inline styles
- pass: No prompt-injection phrasing
- pass: No curl or wget piped into a shell
- pass: No recursive delete of root, home or parent
- pass: No instruction to read or print local credentials
- pass: No base64 blob over 200 characters
- pass: No link to a raw IP address
- pass: No script tag
Source
openqa-cn/codexqa · 8 stars · license Apache-2.0 · pushed 2026-09-23 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_aupfxdbcidry27wx GET https://markdownregistry.com/api/v1/resolve?ref=openqa-cn/codexqa/codexqa-defect-analyzer GET https://markdownregistry.com/api/v1/blob/c4395343b67d92e12cd79430af61653c58c2e1a9d98217095707f867f9e66fed
Your agent does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.
More from openqa-cn/codexqa
Every file in openqa-cn/codexqa