github-health-diagnosis

v1.0.0

GitHub 项目健康度诊断 — 输入仓库地址,AI 自动分析 Issues、代码质量、CI状态,生成诊断报告并发布到飞书 Wiki。

0· 88·0 current·0 all-time

Install

OpenClaw Prompt Flow

Install with OpenClaw

Best for remote or guided setup. Copy the exact prompt, then paste it into OpenClaw for zlszhonglongshen/github-health-diagnosis.

Previewing Install & Setup.
Prompt PreviewInstall & Setup
Install the skill "github-health-diagnosis" (zlszhonglongshen/github-health-diagnosis) from ClawHub.
Skill page: https://clawhub.ai/zlszhonglongshen/github-health-diagnosis
Keep the work scoped to this skill only.
After install, inspect the skill metadata and help me finish setup.
Use only the metadata you can verify from ClawHub; do not invent missing requirements.
Ask before making any broader environment changes.

Command Line

CLI Commands

Use the direct CLI path if you want to install manually and keep every step visible.

OpenClaw CLI

Bare skill slug

openclaw skills install github-health-diagnosis

ClawHub CLI

Package manager switcher

npx clawhub@latest install github-health-diagnosis
Security Scan
VirusTotalVirusTotal
Benign
View report →
OpenClawOpenClaw
Benign
medium confidence
Purpose & Capability
Name/description match the workflow: it collects GitHub data, runs a code-review skill, renders cards, and posts to Feishu. The listed dependencies (github, code-review-skill, card-renderer, feishu-wiki) align with those tasks and explain where credentials and capabilities would come from.
Instruction Scope
Instructions stay within the stated purpose and do not instruct arbitrary data exfiltration or access to unrelated system paths. Minor inconsistencies: README mentions using the 'gh' CLI (gh issue list, gh pr list), but the skill declares no required binaries; Step 1 collects repo metadata but does not explicitly fetch file contents while Step 2 requests code review of files (max_files: 10). It's plausible the 'github' or 'code-review-skill' dependencies handle file retrieval, but this is not spelled out.
Install Mechanism
This is an instruction-only skill with no install spec and no code files executed locally; therefore there is no direct install risk from this package itself.
Credentials
The skill declares no required env vars or credentials, which is reasonable because it delegates to dependent skills. However, practical operation will require GitHub and Feishu credentials (or tokens) held by those dependent skills or supplied as inputs (wiki_space_id/wiki_parent_token). The skill does not request unrelated secrets; confirm dependent skills' credential requirements before use.
Persistence & Privilege
always is false and the skill is user-invocable; it does not request permanent system presence or to modify other skills. No persistence/privilege escalation behavior is specified.
Assessment
This skill appears to do what it says: gather GitHub data, run a code review, generate cards, and publish a Feishu Wiki page. Before installing or running it, verify the following: 1) Where will GitHub and Feishu credentials come from? The skill delegates to other skills (github, feishu-wiki) which likely require tokens—use least-privilege tokens and confirm storage/permissions. 2) Clarify how code files are fetched for review (public repo vs private repo): ensure the code-review-skill or github dependency has explicit, limited access and that you consent to any code being sent to third-party services. 3) The README references the 'gh' CLI but the skill doesn't declare binaries—confirm if the environment needs the GitHub CLI installed or if the github skill handles API access. 4) Test on a public, non-sensitive repository first to observe behavior and outputs. If any dependent skill (code-review-skill, card-renderer, feishu-wiki) is untrusted or opaque, treat that as an additional risk and inspect those skills' credential requirements and runtime behavior before broad use.

Like a lobster shell, security has layers — review code before you run it.

latestvk97fxkvazfx2dsyzztm30ea8c5858q61
88downloads
0stars
1versions
Updated 6d ago
v1.0.0
MIT-0

GitHub 项目健康度诊断助手

一条命令完成:仓库数据抓取 → Issues 分析 → 代码质量审查 → CI/CD 状态检查 → 诊断报告卡片生成 → 飞书 Wiki 发布

核心价值

  • 一键诊断:只需提供一个 GitHub 仓库地址,5分钟内生成完整健康度报告
  • AI 代码审查:集成专业代码审查 Skill,多维度评估代码质量
  • 可视化呈现:生成诊断卡片,团队一眼看懂项目健康状态
  • 知识沉淀:诊断报告自动发布到飞书 Wiki,便于存档和分享

适用场景

  • 技术负责人接新项目,快速了解代码库健康状况
  • 团队代码质量管理,定期生成健康度报告
  • 开源项目维护者定期自检
  • 尽职调查 / Due Diligence 场景下的代码质量评估
  • CI/CD 流水线异常后的根因分析辅助

工作流程(5步)

Step 1 → github
   │   获取仓库基本信息(Stars、Fork、Watch、开放 Issues 数量)
   │   获取最近 30 条 Issues 和 PR 列表
   │   获取最新 Commits 和贡献者信息
   ↓
Step 2 → github (代码审查)
   │   对仓库主要代码文件进行语法/逻辑分析
   │   通过 code-review-skill 执行多维度审查
   ↓
Step 3 → code-review-skill
   │   输出:代码质量评分 / 问题列表 / 安全风险 / 性能建议
   │   汇总为结构化诊断数据
   ↓
Step 4 → card-renderer
   │   生成诊断摘要封面卡(雷达图式健康度展示)
   │   生成详情页卡片(分维度评分 + 问题列表)
   ↓
Step 5 → feishu-wiki
   │   将诊断报告发布为飞书 Wiki 页面
   │   包含:项目概览 / 健康度评分 / 问题详情 / 修复建议

使用方法

触发词

"诊断项目健康度 [仓库地址]"
"帮我看看这个GitHub项目的质量"
"仓库健康检查 [owner/repo]"
"生成项目诊断报告"
"这个项目代码质量怎么样"
"GitHub项目健康度分析"

输入格式

诊断项目健康度 https://github.com/owner/repo
或
诊断项目健康度 owner/repo

输出

  • 飞书 Wiki 页面链接:完整诊断报告(含评分、问题列表、修复建议)
  • 诊断卡片图片:封面卡 + 详情卡,可直接分享

诊断维度(5大维度,满分100)

维度权重评估内容
代码质量30%规范遵循、复杂度、注释覆盖率
Issue 活跃度20%Issue 响应速度、Open Issues 积压量
CI/CD 状态20%Workflow 通过率、测试覆盖率
社区活跃度15%Stars 增长、PR 合并速度、贡献者数量
安全风险15%依赖漏洞、敏感信息泄露风险

健康度评级

  • 🟢 90-100:优秀 — 项目运行良好
  • 🟡 70-89:良好 — 有少量可改进项
  • 🟠 50-69:一般 — 存在较多问题需关注
  • 🔴 0-49:危险 — 需立即介入处理

示例报告结构

📊 GitHub 项目健康度诊断报告
仓库:facebook/react
诊断时间:2026-04-21

🏆 综合评分:92/100 🟢 优秀

各维度评分:
  代码质量      ████████████ 95/100
  Issue 活跃度  █████████░░░ 85/100
  CI/CD 状态    ████████████ 98/100
  社区活跃度    ████████████ 96/100
  安全风险      █████████░░░ 88/100

📋 主要问题:
  1. [中等] 部分组件缺少 TypeScript 类型定义
  2. [低]  少数工具函数缺少 JSDoc 注释

💡 改进建议:
  1. 建议补充 Props 的 PropTypes 或 TypeScript 类型
  2. 建议对核心工具函数添加 JSDoc 注释提升可维护性

Comments

Loading comments...