审计标准

本条目介绍 审计标准 的公开用途。DeepSkill 当前公开目录说明:“审核当前项目是否符合 ~/.claude/CLAUDE.md 中定义的开发标准。将不合规项记录为 GitHub 问题,并编写一份优先级修复计划写入项目的 CLAUDE.md。当用户说“根据设置审核”、“审核标准”、“检查标准合规性”或“审核此项目”时使用。” 适用场景包括测试与安全校验、写代码与代码审查。本页根据公开目录资料和固定版本的上游仓库整理,建议使用前核对该版本的环境、工具、权限及数据处理要求。Skillπ 提供站方用途介绍与原站链接;原始 SKILL.md 和附件仅在对应许可允许时开放。

仓库
jeremylongshore/tons-of-skills-marketplace

安装方式

先用 Prompt 核对来源,也可切换到安装命令或下载可用的安装包。

安装方式

用 AI 助手安装

复制提示词,让助手先核对固定版本和依赖。

查看 Prompt收起 Prompt
先检查当前 Agent 是否已安装 SkillPI Skill Finder。如果尚未安装,请按官方仓库的说明安装并启用:https://github.com/Findskills/skillpi。然后使用 Skill Finder 打开此 SkillPI 详情页并定位 "audit-standards":https://skillpi.com/zh/skills/deepskill-435c6d03c8a347979823a6d496d428c9c3051baa6c928a618d383b142cb397f5。安装前核对上游来源和版本:https://github.com/jeremylongshore/tons-of-skills-marketplace/blob/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/productivity/claude-workflow-skills/skills/audit-standards/SKILL.md。来源不一致时停止并说明差异。核对通过后,用 Skill Finder 安装目标 Skill,并报告安装路径和来源版本。

此命令从 GitHub 来源仓库获取指定技能。运行前请先查看源文件。

运行工具
npx skills add https://github.com/jeremylongshore/tons-of-skills-marketplace/tree/6ea46b08dce892944db7c16f28d0b5308d402402/plugins/productivity/claude-workflow-skills/skills/audit-standards --skill audit-standards
许可协议: —

来源与配置

查看 GitHub 源码

jeremylongshore/tons-of-skills-marketplace固定提交 6ea46b08dce892944db7c16f28d0b5308d402402

配置方式待核实

请查看原始来源中的适用工具、依赖和权限要求。

安装实测待补充

本条目介绍 审计标准 的公开用途。DeepSkill 当前公开目录说明:“审核当前项目是否符合 ~/.claude/CLAUDE.md 中定义的开发标准。将不合规项记录为 GitHub 问题,并编写一份优先级修复计划写入项目的 CLAUDE.md。当用户说“根据设置审核”、“审核标准”、“检查标准合规性”或“审核此项目”时使用。” 适用场景包括测试与安全校验、写代码与代码审查。本页根据公开目录资料和固定版本的上游仓库整理,建议使用前核对该版本的环境、工具、权限及数据处理要求。Skillπ 提供站方用途介绍与原站链接;原始 SKILL.md 和附件仅在对应许可允许时开放。

↓ 0 下载 ↗ 0 分享点击

文件浏览

在原站浏览文件 ↗

原始 SKILL.md

打开原文 ↗
---
name: audit-standards
description: Audits the current project against the development standards defined in ~/.claude/CLAUDE.md. Documents non-compliant findings as GitHub issues and writes a prioritised fix plan to the project CLAUDE.md. Use when the user says audit against settings, audit standards, check standards compliance, or audit this project.
allowed-tools: Read Glob Grep Bash
disable-model-invocation: true
---

# Audit Standards

Project: !`basename $(git rev-parse --show-toplevel 2>/dev/null) 2>/dev/null || basename $PWD`
Branch: !`git branch --show-current 2>/dev/null || echo "unknown"`

Audits the current project against the personal development standards loaded from `~/.claude/CLAUDE.md`.
Generates actionable GitHub issues and a prioritised fix plan.

The development standards from the user's global `~/.claude/CLAUDE.md` are already loaded into context.
The key principles to audit against are:

1. **Codify, Don't Document** — manual steps should be executable scripts
2. **Bash Script UX with pfb** — terminal output uses pfb with correct visual hierarchy
3. **Markdown Standards** — all markdown passes markdownlint with zero warnings
4. **Professional Documentation Tone** — formal docs are objective, not personal
5. **Version Control Everything** — correct files committed, secrets excluded
6. **Fail Fast, Pivot Early** — (process principle, skip for static audit)
7. **Behavioral Integrity** — (process principle, skip for static audit)

## Step 0: Pre-flight check

```bash
gh auth status 2>&1 || { echo "ERROR: gh is not authenticated. Run: gh auth login"; exit 1; }
```

## Step 1: Inventory the project

```bash
find . -not -path './.git/*' -type f | sort
```

Also read:

- `README.md` — documentation tone and completeness
- `CLAUDE.md` — project AI context file exists and is current
- Any `.env`, `.env.template`, `.gitignore` — secrets handling
- Any `*.sh` files — bash scripting standards
- Any `*.md` files — markdown quality

## Step 2: Audit each principle

### Principle 1 — Codify, Don't Document

- Does the README describe manual steps that should instead be scripts?
- Do bash scripts use `#!/usr/bin/env bash` shebang?
- Do scripts have Google-Style header docs (name, description, author, version, usage, dependencies)?
- Do functions have `@param`, `@return`, `@example` doc comments?
- Is author attribution derived from `git config user.name`/`git config user.email`?
- Are scripts idempotent (safe to run multiple times)?

Run: `grep -rn "#!/" --include="*.sh" .` to find all scripts.

### Principle 2 — Bash Script UX with pfb

For each `.sh` file:

- Is pfb used for terminal output (not plain `echo` for status messages)?
- Are emojis passed as a parameter, not embedded in the message string?
- Are log levels (`pfb info/success/warn/error`) used for single-line status only?
- Is visual hierarchy consistent (heading → subheading, not mixed)?

Run: `grep -rn "pfb\|echo" --include="*.sh" .` and compare usage patterns.

### Principle 3 — Markdown Standards

Check for a `.markdownlint.json` at the repo root:

```bash
cat .markdownlint.json 2>/dev/null || echo "MISSING"
```

Run markdownlint if available:

```bash
markdownlint '**/*.md' 2>&1 | head -50
```

Check:

- Blank lines around headings, lists, code blocks, tables
- Code blocks specify a language (no bare triple-backtick fences)
- Prose lines ≤ 120 characters
- URLs wrapped in `<>` or link syntax
- Consistent list numbering (1. 2. 3.)
- YAML frontmatter description fields are single unbroken lines

### Principle 4 — Professional Documentation Tone

Read `README.md` and any formal docs. Flag:

- Second-person language ("you", "your") in formal technical docs (README intro, architecture sections)
- Conversational phrasing in formal sections
- User-specific references that should be generic

Note: READMEs and tutorials may use "you" in instructional sections — this is acceptable.

### Principle 5 — Version Control Everything

```bash
cat .gitignore 2>/dev/null || echo "MISSING .gitignore"
ls .env* 2>/dev/null
```

Check:

- `.gitignore` exists and excludes `.env`, secrets, build artifacts, IDE files
- `.env.template` exists if the project requires environment configuration
- No `.env` with real values present
- AI context files (`CLAUDE*.md`, `AGENTS*.md`, `GEMINI*.md`) are gitignored and symlinked

#### AI context file check

Find all AI context files in the project (excluding `.git`):

```bash
find . -not -path './.git/*' \( -name 'CLAUDE*.md' -o -name 'AGENTS*.md' -o -name 'GEMINI*.md' \) | sort
```

For each file found, verify two things:

1. **Gitignored** — the pattern appears in `.gitignore`:

   ```bash
   git check-ignore -v <file>
   ```

2. **Symlink** — the file is a symbolic link (pointing to the private `ai-context` repo), not a regular tracked file:

   ```bash
   [ -L "<file>" ] && echo "SYMLINK" || echo "REGULAR FILE — should be a symlink"
   ```

A finding is raised for any AI context file that is **not** gitignored OR is **not** a symlink. The expected
pattern is: file is listed in `.gitignore`, stored in the private `ai-context` repo, and symlinked back into
the project directory so local AI tooling finds it normally (see `extract-ai-context.sh` in that repo).

## Step 3: Generate GitHub issues

First, fetch all open issues to avoid creating duplicates:

```bash
gh issue list --state open --limit 100 --json title --jq '.[].title'
```

For each finding, check whether an open issue with a matching title already exists. If one does,
skip it. Only create an issue if no existing open issue covers the same finding.

For each distinct finding with no existing open issue, create a GitHub issue:

```bash
gh issue create \
  --title "<principle>: <brief description of violation>" \
  --body "$(cat <<'EOF'
## Principle

<which standard this relates to>

## Finding

<description of the non-compliance>

## Expected

<what the standard requires>

## Current state

<what was found in the project>

## Suggested fix

<concrete change to make>
EOF
)" \
  --label "enhancement"
```

Use `--label "bug"` for missing required files or broken standards; `--label "enhancement"` for
improvements and style compliance.

Note each issue number as you go.

## Step 4: Write prioritised fix plan to CLAUDE.md

Append or update a section in the project `CLAUDE.md` under the heading
`## Standards Audit — <today's date>`:

```markdown
## Standards Audit — YYYY-MM-DD

Issues generated from `/audit-standards` review against ~/.claude/CLAUDE.md.
Suggested fix order:

### Group 1 — Correctness (fix first)

- #N: <title>
- #N: <title>

### Group 2 — Standards Compliance

- #N: <title>
- #N: <title>

### Group 3 — Quality Improvements

- #N: <title>
- #N: <title>
```

Order: missing required files first, then standards violations, then style improvements.

## Step 5: Report summary

Output a brief summary:

- Total issues created (with links)
- Which principle had the most findings
- Top-priority fix
- Link to the CLAUDE.md section added

当前收录版本

0.0.1

核对来源版本 ↗

质量评测

满分 100。前四项来自 DeepSkill 收录的质量报告。

场景价值85/100
评分依据

提供完整的项目标准合规性审核流程

触发精准90/100
评分依据

触发词覆盖全面,包括多种审核相关表达

结构规范85/100
评分依据

步骤划分清晰,包含完整的执行流程

执行可靠88/100
评分依据

工具来源明确,使用标准的命令行工具

证据可验 (前四项均值)87/100
计算方式

取前四项原始分值的算术平均,四舍五入取整。可通过收录报告核对各项评分与说明,运行记录需单独查证。

权限与使用前检查

根据当前主文档自动整理,命中内容可能是示例或防护说明;不是实际权限声明,也未覆盖附件、外部依赖或运行行为。

未从主文档识别到上述常见操作,不能据此判断无需权限或没有风险。

本站展示提示不会授予权限。涉及文件改动、密钥、外部发送或费用时,仍由你使用的工具按需确认。

安全评级与访问范围

risk · 静态证据核验,不代表运行安全保证。

使用前请由宿主核对并授权实际文件读写、外部服务和联网目的地;本站的核验不会授予任何运行权限。

来源与使用声明 SkillPi 的部分技能资料来自公开来源或创作者提交。第三方内容的权利归各自权利人所有。安装或运行前,请查看原站的许可、权限要求和使用风险。

来源与使用

请按详情页的作者、原始来源和许可证核对使用条件,运行前检查技能需要的权限。目录收录不等于已验证安装或安全性。