model-prompt-audit SKILL.md skillA
model-prompt-audit from hanzhangzzz/agent-skills-zh is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.
mdr add hanzhangzzz/agent-skills-zh/model-prompt-audit@git:20261028.bd6709dPins git:20261028.bd6709d in mdr.lock. Pin the fingerprint instead.
- Audit grade
- A, 17 of 17 checks passed
- Kind
- SKILL.md skill
- Repository
- hanzhangzzz/agent-skills-zh
- Path
model-prompt-audit/SKILL.md- Versions recorded
- 1
- Latest version
- git:20261028.bd6709d, committed Oct 28, 2026
- Size
- 7,225 bytes
- Repository stars
- 1
- License
- MIT
Audit the global agent configuration (CLAUDE.md / AGENTS.md, rules, settings, behavioral plugins) against the current official prompting guide of the Claude model in use. Fetches the guides live at run time instead of relying on memory, derives how prompting paradigms shifted across model generations, flags anti-patterns, duplicates, stale rules and unused patterns, proposes changes, and applies them only after approval. Use when the user says /model-prompt-audit, 提示词范式审计, 我的 CLAUDE.md 过时了吗…
Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.
What the file says
# model-prompt-audit 模型换代后，写给上一代模型的规则会变成负担：曾经需要的"多想一步""再检查一遍"在新模型上变成过度验证，为旧缺陷写的禁令在新模型上压住了它本来会做对的事。本 skill 用**当前**官方文档做一次差距审计：范式怎么变了、你的配置里哪些已经反范式、哪些新范式还没用上，然后经你批准再改。 ## 铁律 - **文档现取，不用记忆。** 模型指引每次发版都在变，训练记忆里的"最佳实践"很可能就是要被清理的旧范式。所有结论必须能指向本次抓取到的文档段落；抓取失败就停下报告，不用记忆补位。 - **先提案后动手。** 盘点与判定阶段只读；任何文件改动、插件启停、设置修改都在用户批准之后，且改前备份。 - **只判模型行为类指令。** 领域规则（时区、身份、危险操作清单、真实事故教训）不因为写成"禁止"就算反范式；要审的是那些试图矫正模型行为的指令。 ## 执行步骤 ### 0. 前置检查 1. `python3 --version` 可用；能访问文档站（`curl -sI https://platform.claude.com/llms.txt` 返回 200）。任一失败，报告并停止。 2. 确定当前模型 ID，优先级：用户参数 > 宿主设置文件的 `model` 字段（Claude Code 为 `~/.claude/settings.json`，可带 `[1m]` 后缀）> 会话自述。用会话自述时明确标注"未从设置核实"。 ### 1. 抓取当前文档 ```bash SKILL_DIR="$(dirname "$(readlink -f "$0")")" # 在 SKILL.md 相邻目录取脚本，不假设安装路径 OUT="$(mktemp -d -t model-prompt-audit-XXXXXX)" python3 "$SKILL_DIR/scripts/fetch_guides.py" --model <model-id> --out "$OUT" ``` 脚本从文档站的 `llms.txt` 索引挑出：所有 `prompting-claude-*` 模型指引、跨模型的 best-practices 页、当前模型家族的 what's new 与 migration guide，下载为 Markdown 并写 `index.json`。退出码 1 表示没找到当前模型自己的指引：改用同家族最近的指引并在报告里注明，不要假装它就是当前模型的。 ### 2. 建立谱系并通读 1. 读当前模型的指引，找出它对比的上一代（开头通常写 "differences from X"），再读 X 的指引，再往前一代。至少两跳；指引总数不多时全读。 …
Read the whole file at its exact version.
How to install
mdr add hanzhangzzz/agent-skills-zh/model-prompt-audit@sha256:0285b13de06ebbbfThe line at the top of this page pins the label git:20261028.bd6709d, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.
[](https://markdownregistry.com/a/art_shxh2wg2m2i4tm6t)
3 badge image fetches in 30 days. What the badge shows.
Versions
The registry has recorded 1 version of this file, committed on Oct 28, 2026.
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 (7225 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
hanzhangzzz/agent-skills-zh · 1 stars · license MIT · pushed 2026-10-29 · branch main
API
GET https://markdownregistry.com/api/v1/artifacts/art_shxh2wg2m2i4tm6t GET https://markdownregistry.com/api/v1/resolve?ref=hanzhangzzz/agent-skills-zh/model-prompt-audit GET https://markdownregistry.com/api/v1/blob/0285b13de06ebbbfd648857f88f0804c20019a62bde7378a38323b8019239ae5
Your AI assistant 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.
Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.