Token导航 LogoToken导航

Agent Skill · 开发

github-chat-ops(GitHub 聊天 OPS)

用于围绕 GitHub 仓库、Issue、Pull Request、分支、提交和代码协作流程提供辅助能力。它适合让 Agent 查询项目状态、整理变更、辅助创建或检查协作事项,并把仓库中的信息转成可执行的下一步。使用时需要区分只读查询和写入操作;涉及创建 PR、修改 Issue、推送分支或访问私有仓库时,应确认 token 权限、目标仓库范围和用户授权。

github-chat-ops 是开发类 Agent Skill,可用宿主为 OpenClaw,通过ClawHub获取,权限为接触敏感数据;许可证为 MIT-0,第三方扫描有提醒项,按安装量在开发类的 21,746 个 Skill 里排第 1,457。

总安装 29,095周安装 1,177GitHub Stars 公开资料未说明下载 9,134核验 2026-05-01

安装说明

本站只整理中文说明,不托管安装包,也不代用户安装。
GitHub
许可证
MIT-0
版本
1.0.0

安装方式

通过对话安装

复制提示词发给支持本地命令或 Skills 的 AI 助手,先确认命令和权限,再让它执行。

请帮我安装这个 Agent Skill:github-chat-ops(GitHub 聊天 OPS)
来源仓库:https://github.com/iamkalio/github-chat-ops
安装命令:
openclaw skills install github-chat-ops
安装前请先检查当前环境是否支持对应 CLI,并向我确认将要执行的命令、安装目录、联网范围和文件读写权限;确认后再执行。

命令行安装

复制命令到本机终端执行。该命令会通过 OpenClaw 从第三方来源获取 Skill;本站只展示命令,不托管安装包,也不自动执行。

ClawHubOpenClaw
openclaw skills install github-chat-ops

这个 Skill 是做什么的

开发类 Skill 面向写代码、改代码和排查问题,一般会读写仓库文件,部分还会执行构建或测试命令。

通过聊天为非技术请求者管理单个 GitHub 存储库 - 在他们共享存储库 URL 和临时个人令牌后,提取状态,总结谁在何时做了什么,并直接通过 GitHub API 创建/跟踪问题。

上手步骤

4 步
  1. 1

    确认自己的宿主工具

    公开资料标注支持 OpenClaw;先确认自己用的是其中之一,再按对应工具的 Skills 目录放置。

  2. 2

    拿到安装命令

    可以直接复制的命令是 openclaw skills install github-chat-ops;命令里的占位符换成自己的路径或密钥再执行。

  3. 3

    安装后重启宿主

    命令执行完要完全退出宿主工具再打开,让新装的 Skill 被重新扫描到,否则列表里可能看不到。

  4. 4

    按最小权限试用

    它需要接触敏感数据,先在测试目录或测试账号里跑,确认没问题再放到正式项目。

常见问题

6 条

github-chat-ops 怎么安装?

在终端执行 openclaw skills install github-chat-ops。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

github-chat-ops 支持哪些 AI 工具?

公开资料标注支持 OpenClaw。其他工具只要能读取本地 Skill 目录,一般也能手动放进去,但要自己确认触发方式是否一致。

github-chat-ops 需要什么权限?

权限级别是接触敏感数据。该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

github-chat-ops 是免费的吗?

来源标注的许可证是 MIT-0。本站不代收任何费用,也不托管安装包;能否商用、要不要署名,以许可证原文为准。

github-chat-ops 用的人多吗?

公开数据是累计安装 29,095 次,近一周安装 1,177 次,下载 9,134 次。这些数字来自第三方目录统计,只能当作热度参考,不代表质量或安全性。

github-chat-ops 安全吗?

第三方扫描有提醒项,说明存在需要人工确认的行为。安装前请打开源码看清它读了什么、联网到哪里。

值得选的地方和要先想清楚的地方

值得选的地方

  • 专注 OpenClaw 一种宿主,触发方式和使用习惯不用来回切换。
  • 许可证明确为 MIT-0,商用和分发边界好判断。

先想清楚的地方

  • 需要接触敏感数据,装到常用环境前要先评估影响面。
  • 第三方扫描存在提醒项或未通过项,需要人工逐条确认。
  • 第三方目录整理,版本更新和行为细节以来源仓库为准。

适不适合你

更适合

  • 已经在用 OpenClaw,想直接扩展它能力的人。
  • 需要处理开发类任务、又不想每次都手写提示词的人。

不太适合

  • 环境里有大量密钥、凭证,不方便给第三方 Skill 读取的人。
  • 需要厂商级技术支持和 SLA 的生产系统。

开发类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
skill-vetter

技能审查员 · OpenClaw

8,779,018986操作浏览器MIT-0通过
ontology

本体论 · OpenClaw

6,636,960576只读MIT-0通过
entra-app-registration

内部应用程序注册 · Codex、Claude

6,631,872764接触敏感数据未明确通过
entra-app-registration

内部应用程序注册 · Codex、Claude

3,757,680201接触敏感数据未明确通过
nano-banana-pro

纳米香蕉专业版 · OpenClaw

3,526,898364接触敏感数据MIT-0有提醒
github-chat-ops

GitHub 聊天 OPS · 本页

29,095—接触敏感数据MIT-0有提醒

按安装量取同分类前几名;数字来自第三方目录统计,只作热度参考,不代表质量结论。

SKILL.md 原文

name
github-chat-ops
description
Manage a single GitHub repository via chat for non-technical requesters—after they share the repo URL and a temporary personal token, pull status, summarize who did what and when, and create/follow up on issues directly through the GitHub API.

Overview

Use this skill whenever a non-technical person (often over WhatsApp) needs lightweight GitHub help without cloning or forking a repo. Typical asks:

  • "Tell me what changed recently and who did it."
  • "Create an issue describing X."
  • "Follow up on existing issues or PRs."

The workflow relies entirely on the GitHub REST API using a personal access token (PAT) the requester provides during the chat.

1. Gather prerequisites every session

  1. Repo identifier – ask for full URL or owner/name.
  2. PAT – confirm it has repo scope (private repos) or public_repo (public). Remind them to generate a short-lived token and send it in the chat; you’ll discard it afterward.
  3. Task brief – what they need (issue, summary, follow-up) plus timeframe (e.g., "last 7 days").

_Always restate the inputs back to them before acting. If anything is missing, pause and ask._

2. Handle tokens safely

  • Paste the token into a temporary shell variable in the current session only:
  export GITHUB_TOKEN="<token-from-chat>"
  • Never save tokens to disk or log files. When finished, run unset GITHUB_TOKEN.
  • Every API call must set Authorization: Bearer $GITHUB_TOKEN and Accept: application/vnd.github+json.

3. Fetch repo context before acting

  1. Health check: GET /repos/{owner}/{repo} – confirms access and surfaces default branch.
  2. Recent commits (for summaries / who-did-what):

- GET /repos/{owner}/{repo}/commits?since=<ISO8601>&until=<ISO8601> - For per-author breakdown, add author=<username> or group results locally.

  1. Issues & PRs: GET /repos/{owner}/{repo}/issues?state=all&since=<ISO8601>.

- Distinguish PRs via pull_request key.

Record the raw JSON responses (e.g., save to /tmp/commits.json) if you need to run jq filters before summarizing.

4. Deep repo inspection without cloning

When you need file-level context (to quote code in an issue or explain why a commit matters), walk the tree via the REST API:

  1. List directories/files: GET /repos/{owner}/{repo}/contents/<path>?ref=<branch> returns metadata plus download URLs.
  2. Fetch raw blobs: reuse the download_url or call GET /repos/{owner}/{repo}/contents/<path> with header Accept: application/vnd.github.raw.
  3. Large trees: GET /repos/{owner}/{repo}/git/trees/<sha>?recursive=1 to grab the whole structure, then request the files you care about.
  4. Cache what you read per session (e.g., store under /tmp/github-chat-ops/<repo>/...) so subsequent lookups avoid extra API calls.

Always mention the file + path + relevant snippet when writing summaries or issues.

5. Summaries for non-technical readers (with real changes)

Translate activity into plain language:

  • Group commits by contributor, then describe what actually changed (features/tests/configs) rather than just commit titles.
  • For each commit, hit GET /repos/{owner}/{repo}/commits/{sha} to see files[] (filenames, additions/removals, patch).
  • Extract the top 1–2 bullets per contributor: e.g., "Updated quiz_generator.py to support context prompts and added 3 YAML fixtures."
  • Mention timestamps in the user’s timezone (Africa/Lagos unless told otherwise).
  • Highlight status (merged, open, blocked) and outstanding follow-ups.

Keep summaries short, bullet-style, and avoid jargon.

6. Creating or updating issues

  1. Clarify the problem, expected outcome, priority, and owners while chatting.
  2. Draft the issue body locally in Markdown (include background, repro steps, acceptance criteria).
  3. Create the issue via POST /repos/{owner}/{repo}/issues with JSON payload:
   {
     "title": "...",
     "body": "...",
     "assignees": ["username"],
     "labels": ["priority:high"]
   }
  1. Echo the created issue URL back to the user and summarize what was filed.

_For follow-ups_, use PATCH /repos/{owner}/{repo}/issues/{number} to update state or assignees, and POST /repos/{owner}/{repo}/issues/{number}/comments for status notes.

7. Conversation pattern (WhatsApp-ready)

  1. Confirm: "I’ll need the repo URL + a temporary token with repo scope—can you share those now?"
  2. After each action, report in natural language first, then share links/details.
  3. Keep tokens out of summaries. If the user sends screenshots or sensitive info, acknowledge and avoid re-sharing it elsewhere.

8. Daily automation & cron jobs

  • Store long-lived secrets outside the skill (e.g., .env.github-chat-ops with GITHUB_CHAT_OPS_TOKEN, repo, timezone).
  • Build reusable scripts under scripts/ (see scripts/github_chat_ops_daily.py) that load env vars, call the same APIs, and print a ready-to-send summary.
  • When scheduling via cron, run the script from the workspace root, capture stdout verbatim for the message, and surface errors if the script exits non-zero.
  • Let end users customize repo/timezone by editing the env file; document that in the skill release notes so downstream consumers know where to look.

9. References

Use references/github-api-cheatsheet.md for ready-made curl templates covering the endpoints above plus pagination tips.

平台分布

OpenClaw

91.33%
按下载量换算8,342

安全审计

VirusTotal

可疑

ClawScan

可疑

Static analysis

未展示

权限和风险

接触敏感数据

该 Skill 可能接触密钥、Token、环境变量或敏感配置,应进入高风险复核队列,默认不自动发布。

安装前确认

本站仅展示第三方公开信息,不托管安装包,不提供自动安装或运行环境。安装前应自行审查源码、依赖和命令行为。来源安全扫描存在提醒或未通过项,安装前请自行复核。目前只有一个公开来源,建议同时核对源仓库。

继续浏览同类 Skills