SkillFoxx

GitHub Explorer skill

An OpenClaw skill that analyzes a GitHub project across many sources: gist, health, top issues, rivals and community view

  • 79 ★ on the repository
  • Last commit
  • MIT
Skill

Extended access

Why this level

  • The skill visits external sites and search services and may open pages in the agent browser
  • Requests to Brave, Exa and Tavily spend your keys if dependencies are installed
All reasons and checks

blessonism/github-explorer-skillAuthor:blessonism

Install

Works in

In your terminal, with SkillFoxx CLI

npx skillfoxx add skills/github-explorer

Detects the agents on your machine, shows the access level and pins the version.

Other ways to install

Access and checks

How we review

Extended accessThe tool runs code or makes network calls and only reads external data.

  • The skill visits external sites and search services and may open pages in the agent browser
  • Requests to Brave, Exa and Tavily spend your keys if dependencies are installed

This is third-party code. Review the repository files before installing.

What it does

GitHub Explorer is a skill for the OpenClaw agent that studies a GitHub project deeper than reading the README. It collects data from issues and commits, from Chinese platforms such as Zhihu, V2EX and WeChat, from Medium and Dev.to technical blogs and from Twitter discussions. It then classifies the project stage: early experiment, fast growth, mature or stalled, and picks the most substantive issues. The result comes in a fixed template: positioning, mechanism, project health by stars, forks, license and commit trend, top issues, fitting and unfitting scenarios, rival comparison, presence in DeepWiki and Zread.ai, demo, papers, community voice and a personal verdict. Every quote carries its source link.

Who it is for. Developers and tech leads choosing a library or tool who want a quick project summary with community opinion, Chinese included.

Good fit when

  • You choose between two projects and want a rival comparison with links
  • Before adoption you need to know whether a project is alive and see its stage: growth, mature or stalled
  • You are curious what Zhihu, V2EX and Twitter say about a project, with quotes and URLs
  • Among thousands of issues you need the three to five most important and the gist of the debate
  • You want to check whether a project is in DeepWiki and Zread.ai and has an arXiv paper

Not a fit when

  • The agent is not OpenClaw: the skill calls its built-in `web_search`, `web_fetch` and `browser`
  • You need a file-by-file source code review: the skill evaluates a project from outside, by its surroundings
  • The agent network is closed: collection goes to external platforms
  • The project is private with no public discussion: there will be no data for a summary

Example request

Analyze the microsoft/graphrag project: stage, health, best issues, rivals and what communities say about it

Limitations

The skill is written for OpenClaw and calls its built-in `web_search`, `web_fetch` and `browser` tools. Extended search comes from separate skills in the `openclaw-search-skills` repository: `search-layer` with Brave, Exa and Tavily, `content-extract` for platforms that resist scraping and `mineru-extract`. Without them github-explorer falls back to built-in tools and core functions remain. Search needs keys for the respective services. The author notes that the repository is included in the aggregating `openclaw-skills`. The README and reports are in Chinese.

How to disable. Delete the `github-explorer` folder from `~/.openclaw/skills/` or from your agent skills directory. Dependency skills from `openclaw-search-skills` are removed separately.

FAQ

How do I install GitHub Explorer?

Run `npx skills add https://github.com/blessonism/github-explorer-skill`. Manually: `git clone https://github.com/blessonism/github-explorer-skill.git github-explorer` into `~/.openclaw/skills/`.

Which dependencies are needed?

None are required: the skill uses built-in `web_search`, `web_fetch` and `browser`. Results are better with `search-layer`, `content-extract` and `mineru-extract` from the `openclaw-search-skills` repository.

How do I start an analysis?

Tell the agent, for example "analyze https://github.com/microsoft/graphrag" or "look at the langchain project". The skill triggers itself, collects data and prints a report.

What does the report contain?

Positioning, mechanism, project health, top issues, scenarios, limitations, rivals, DeepWiki presence, demo, papers, community voice and a verdict. The report title links to the repository.

How does the skill pick a search strategy?

By intent: project research, rival comparison and fresh news checks use different queries. Results are ranked by source authority, freshness and keyword match.

Discussion

No questions or reviews yet.

Skills

A skill that dissects an open-source project's architecture and returns one Markdown report with Mermaid diagrams and a comparison with rivals

SkillExtended access503

Skill that judges a GitHub project like an outside contributor: how outside PRs get merged, how fast maintainers reply.

SkillExtended access93

OpenClaw Agent Skills

openclaw-agent-skills

Official

Shared skills for coding agents that work on OpenClaw projects: review and other workflows. A narrow topic for people who write OpenClaw code

SkillExtended access1.1K

A code review skill with four independent probes, evidence-backed findings and a merge gate derived from JSON

SkillExtended access479

Twenty skills for reading unfamiliar code, reviewing documents and commits, translating Markdown and OpenSpec-driven development

SkillExtended access92

README in short

The README lists skill traits: multi-source collection, project stage assessment, intent-aware search, rival comparison and a fixed report template. It then covers installation, dependencies, example prompts and working principles: source links, a fallback collection mode and concrete quotes instead of vague ratings.

README badge

Original SKILL.md

---
name: github-explorer
description: >
  Deep-dive analysis of GitHub projects. Use when the user mentions a GitHub repo/project name
  and wants to understand it — triggered by phrases like "帮我看看这个项目", "了解一下 XXX",
  "这个项目怎么样", "分析一下 repo", or any request to explore/evaluate a GitHub project.
  Covers architecture, community health, competitive landscape, and cross-platform knowledge sources.
---

# GitHub Explorer — 项目深度分析

> **Philosophy**: README 只是门面，真正的价值藏在 Issues、Commits 和社区讨论里。

## Workflow

```
[项目名] → [1. 定位 Repo] → [2. 多源采集] → [3. 分析研判] → [4. 结构化输出]
```

### Phase 1: 定位 Repo

- 用 `web_search` 搜索 `site:github.com <project_name>` 确认完整 org/repo
- 用 `search-layer`（Deep 模式 + 意图感知）补充获取社区链接和非 GitHub 资源：
  ```bash
  python3 skills/search-layer/scripts/search.py \
    --queries "<project_name> review" "<project_name> 评测 使用体验" \
    --mode deep --intent exploratory --num 5
  ```
- 用 `web_fetch` 抓取 repo 主页获取基础信息（README、Stars、Forks、License、最近更新）

### Phase 2: 多源采集（并行）

**⚠️ GitHub 页面抓取规则（强制）**：GitHub repo 页面是 SPA（客户端渲染），`web_fetch` 只能拿到导航栏壳子，**禁止用 web_fetch 抓 github.com 的 repo 页面**。一律使用 GitHub API：
- README: `curl -s -H "Authorization: token {PAT}" -H "Accept: application/vnd.github.v3.raw" "https://api.github.com/repos/{owner}/{repo}/readme"`
- Repo 元数据: `curl -s -H "Authorization: token {PAT}" "https://api.github.com/repos/{owner}/{repo}"`
- Issues: `curl -s -H "Authorization: token {PAT}" "https://api.github.com/repos/{owner}/{repo}/issues?state=all&sort=comments&per_page=10"`
- Commits: `curl -s -H "Authorization: token {PAT}" "https://api.github.com/repos/{owner}/{repo}/commits?per_page=10"`
- File tree: `curl -s -H "Authorization: token {PAT}" "https://api.github.com/repos/{owner}/{repo}/git/trees/{branch}?recursive=1"`

PAT 见 TOOLS.md。

以下来源**按需检查**，有则采集，无则跳过：

| 来源 | URL 模式 | 采集内容 | 建议工具 |
|---|---|---|---|

Report a problem