Token导航 LogoToken导航TokenDH.com
开发操作浏览器clawhub未标认证来源可访问clear审计通过

openclaw-github-copilotOpenClaw GitHub GitHub Copilot 开发

Agent Skill

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。

总安装

4,602

周安装

188

GitHub Stars

1

下载量

1,489
OpenClaw

安装说明

本站只整理中文说明和来源信息,不托管安装包,也不代用户安装。

GitHub

来源数

2

许可证

MIT-0

最后核验

2026-05-01

来源状态

来源可访问

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:openclaw-github-copilot(OpenClaw GitHub GitHub Copilot 开发)
来源仓库:https://github.com/yanglinzhen/openclaw-github-copilot
安装命令:
openclaw skills install openclaw-github-copilot
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 openclaw-github-copilot

简介

通过 copilot-bridge 集成 GitHub Copilot,作为 OpenClaw 的编码代理提升开发效率。

  • 适合程序员在 OpenClaw 环境中获得智能代码建议与补全,支持多语言项目。
  • 首次设置需登录 Copilot 账号并切换默认模型,后续自动注入提示到编码上下文。
  • 依赖 npm 包与本地 Node 环境,需同意 Copilot 服务条款并遵守使用配额。
  • 结果仅供参考,关键代码变更前请人工审查,避免过度依赖导致逻辑错误。

SKILL.md

name
openclaw-github-copilot
description
Use GitHub Copilot as your OpenClaw coding agent via the built-in copilot-bridge. Use when setting up Copilot for the first time, switching the default model, diagnosing auth or alias issues, or giving someone the shortest working path to copilot-bridge/github-copilot.
homepage
https://docs.openclaw.ai/cli/models
metadata
{"openclaw":{"emoji":"🐙","homepage":"https://docs.openclaw.ai/cli/models","requires":{"bins":["openclaw"]},"install":[{"id":"node","kind":"node","package":"openclaw","bins":["openclaw"],"label":"Install OpenClaw CLI"}]}}

OpenClaw GitHub Copilot

Use this skill to make GitHub Copilot the simplest possible OpenClaw default.

This skill is for four jobs:

  1. first-time GitHub Copilot setup in OpenClaw
  2. switching the default model to Copilot
  3. fixing common copilot-bridge/github-copilot auth or alias issues
  4. giving the user one short command path instead of a pile of manual steps

What this skill assumes

  • openclaw is installed
  • the user has GitHub Copilot access on the GitHub account they will log in with
  • the agent prefers OpenClaw CLI commands over hand-editing config files

Fast path

Start with the single diagnostic command:

bash {baseDir}/scripts/copilot-quickstart.sh --probe

If the user wants Copilot to become the default model and auth is already valid:

bash {baseDir}/scripts/copilot-quickstart.sh --probe --activate

If auth is missing or expired and you are in an interactive TTY:

bash {baseDir}/scripts/copilot-quickstart.sh --probe --login --activate

That wrapper is the preferred entrypoint because it:

  • checks whether the Copilot model exists
  • checks whether the short alias exists
  • optionally creates the alias when asked
  • optionally runs the official GitHub Copilot device login
  • optionally switches the default model to Copilot

Canonical identifiers

  • Full model id: copilot-bridge/github-copilot
  • Preferred alias: copilot-auto

Always verify what OpenClaw already knows before claiming anything:

openclaw models list --plain
openclaw models aliases list
openclaw models status --plain

If the alias exists, prefer it in human-facing instructions because it is shorter:

openclaw models set copilot-auto

If the alias is missing, use the full model id:

openclaw models set copilot-bridge/github-copilot

Recommended workflow

1. Diagnose first

Run:

bash {baseDir}/scripts/copilot-status.sh

Use the stronger live probe when auth freshness matters:

bash {baseDir}/scripts/copilot-status.sh --probe

The script reports:

  • whether the Copilot model is registered
  • whether copilot-auto points at that model
  • what the current default model is
  • whether GitHub Copilot is already the active default

2. Authenticate only with the official flow

If the model exists but auth is missing or stale, use:

openclaw models auth login-github-copilot

Important:

  • this requires an interactive TTY
  • it may require a browser/device-code step
  • do not pretend it can be completed silently in the background

3. Switch the default model

If the user explicitly wants Copilot as the default model:

bash {baseDir}/scripts/copilot-activate.sh

Or use the wrapper:

bash {baseDir}/scripts/copilot-quickstart.sh --activate

4. Confirm result

Verify with:

openclaw models status --plain

The configured default should resolve to copilot-bridge/github-copilot.

If the current session still behaves like the old model, start a fresh OpenClaw session. Skill and default-model changes are most reliable in a new session.

Good trigger phrases

Use this skill for requests like:

  • "让 OpenClaw 用 GitHub Copilot"
  • "把默认 agent 切到 Copilot"
  • "检查 Copilot bridge 配好了没有"
  • "GitHub Copilot 在 OpenClaw 里怎么登录"
  • "给我一个最简单的 Copilot 配置流程"
  • "copilot-auto 不生效"

Practical response pattern

When helping a user, keep the answer operational:

  1. say whether the model exists
  2. say whether it is already the default
  3. give the next exact command
  4. only mention manual alias cleanup or config editing if the CLI flow is insufficient

Example:

GitHub Copilot 这个模型已经在 OpenClaw 里了,但当前默认模型不是它。
直接执行 `bash {baseDir}/scripts/copilot-quickstart.sh --probe --activate`。
如果后面提示鉴权失效,再跑 `openclaw models auth login-github-copilot`。

References

  • references/setup-guide.md - full first-time setup walkthrough
  • references/troubleshooting.md - missing model, alias, auth, and session behavior
  • references/publish-checklist.md - release checklist for publishing to ClawHub

适合场景

01

OpenClaw 用户查找和安装 Skill 时

02

用户想查找某类 Agent Skill 时

03

需要根据任务场景推荐可安装能力包时

04

需要对比不同来源的安装命令和来源信息时

能力概览

能力 1

按任务关键词查找相关 Skills

能力 2

展示可复制的安装命令

能力 3

保留来源站点、仓库和原始说明,方便继续核验

能力 4

补充不同宿主或平台的使用分布数据

能力 5

展示第三方安全扫描或审计结果

安装后应在对应宿主中按原始 README 的触发条件使用;具体调用方式请以来源页面和 README 为准。

平台分布

OpenClaw

81.02%
按下载量换算1,206

安全审计

VirusTotal

通过

ClawScan

通过

Static analysis

通过

权限和风险

操作浏览器

该 Skill 可能涉及浏览器控制能力,使用时可能读取或操作网页内容,需要在受控环境中确认权限边界。

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。当前只有一个来源,正式发布前建议补源仓库或其他目录站核验。

来源信息

继续浏览同类 Skills