Token导航 LogoToken导航

Agent Skill · 研究检索

sync-to-copilot(sync TO GitHub Copilot 搜索)

sync-to-copilot 用于查找、检索和筛选相关信息,适合在 Codex、Claude、Cursor、Gemini CLI 中需要根据关键词、任务场景或来源线索快速定位候选结果时使用。可结合来源仓库、安装命令和原始 README 继续核验具体用法。安装前建议确认权限范围、维护状态,以及是否会触发联网、命令执行或文件读写。

sync-to-copilot 是研究检索类 Agent Skill,可用宿主为 Codex、Claude、Cursor、Gemini CLI,通过GitHub 仓库获取,权限为只读;许可证信息还不明确,第三方扫描已通过,按安装量在研究检索类的 73,675 个 Skill 里排第 66,577。

总安装 261周安装 11GitHub Stars 6下载 92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
unknown

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:sync-to-copilot(sync TO GitHub Copilot 搜索)
来源仓库:https://github.com/duc01226/easyplatform
仓库路径:skills/sync-to-copilot
安装命令:
npx skills add https://github.com/duc01226/easyplatform --skill sync-to-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 npx skills 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。

skills.shnpx skills
npx skills add https://github.com/duc01226/easyplatform --skill sync-to-copilot

这个 Skill 是做什么的

研究检索类 Skill 主要解决「找资料、核对事实、把来源整理清楚」的问题,通常会让助手先检索再总结。

用于在 Codex、Claude、Cursor、Gemini CLI 中查找和筛选相关信息。支持根据关键词、任务场景或来源线索快速定位候选结果。可结合原始 README 和仓库内容进一步核验具体用法。安装前建议确认权限范围和维护状态,避免触发不必要操作。安装方式:通过 GitHub 仓库使用 npx 命令添加。

来源仓库内路径:skills/sync-to-copilot

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 Codex、Claude、Cursor、Gemini CLI;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 npx skills add https://github.com/duc01226/easyplatform --skill sync-to-copilot;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

    它标注为只读,先用只读任务试一次,确认行为符合预期再扩大使用范围。

常见问题

6 条

sync-to-copilot 怎么安装?

在终端执行 npx skills add https://github.com/duc01226/easyplatform --skill sync-to-copilot。装完完全退出宿主工具再打开,让它重新扫描本地 Skill。

sync-to-copilot 支持哪些 AI 工具?

公开资料标注支持 Codex、Claude、Cursor、Gemini CLI。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。

sync-to-copilot 需要什么权限?

权限级别是只读。该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

sync-to-copilot 是免费的吗?

公开资料没有标注明确许可证,收费和商用限制都无法确认。正式使用前请到来源仓库核对授权说明。

sync-to-copilot 用的人多吗?

公开数据是累计安装 261 次,近一周安装 11 次,下载 92 次,GitHub 6 星。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。

sync-to-copilot 安全吗?

第三方扫描结果为通过,但这只覆盖已知的静态检查项,不能替代自己读一遍源码和安装命令。

值得选的地方和要先想清楚的地方

值得选的地方

  • 同时适配 4 种工具,换客户端不用重新找替代品。
  • 第三方扫描已通过,常见的高风险行为没有命中。
  • 标注为只读权限,试用阶段不容易误改本地文件。

先想清楚的地方

  • 许可证没有明确标注,能不能商用、能不能再分发都不确定。
  • 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 Codex,想直接扩展它能力的人。
  • 需要处理研究检索类任务、又不想每次都手写提示词的人。
  • 只想先试试效果、不希望它改动本地文件的人。

不太适合

  • 需要明确授权才能用到公司项目里的场景。
  • 需要厂商级技术支持和 SLA 的生产系统。

研究检索类 Skill 对比

6 个同类
Skill总安装GitHub 星权限许可证扫描
find-skills

寻找技能 · Codex、Claude

31,199,81316,486操作浏览器MIT有提醒
self-improving-agent

自我改进 Agent · OpenClaw

16,464,0023,396接触敏感数据MIT-0通过
soultrace

灵魂追踪 · Codex、Claude

7,588,94412需要联网未明确有提醒
self-improving

自我完善 · OpenClaw

6,829,2841,046需要联网MIT-0通过
microsoft-foundry

微软代工厂 · Codex、Claude

6,614,144764调用外部服务未明确有提醒
sync-to-copilot

sync TO GitHub Copilot 搜索 · 本页

2616只读未明确通过

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

Skill: sync-to-copilot

Sync Claude Code knowledge to GitHub Copilot instructions. Two-tier output: project-specific + common protocol.


Quick Summary

Purpose: Keep Copilot instructions in sync with Claude Code workflows, dev rules, and project-reference docs.

Architecture (Two-Tier):

  1. .github/copilot-instructions.md — Project-specific (always loaded by Copilot)

- TL;DR golden rules, decision table - Project-reference docs index with READ prompts - Key file locations, dev commands

  1. .github/instructions/common-protocol.instructions.md — Generic protocols (applyTo: **/*)

- Prompt protocol, before-editing rules - Workflow catalog (from workflows.json) - Workflow execution protocol - Development rules (from development-rules.md)

  1. .github/instructions/{group}.instructions.md — Per-group (applyTo: file patterns)

- Enhanced summaries per doc with READ prompts - Groups: backend, frontend, styling, testing, project

What gets synced:

  • Workflow catalog (from workflows.json) — SCRIPT-GENERATED
  • Dev rules (from development-rules.md) — SCRIPT-GENERATED
  • Project-reference summaries (from copilot-registry.json) — SCRIPT-GENERATED
  • Enriched section headings and key patterns — AI-GENERATED (this skill)

Usage:

/sync-to-copilot

Script: .claude/scripts/sync-copilot-workflows.cjs


Be skeptical. Apply critical thinking, sequential thinking. Every claim needs traced proof, confidence percentages (Idea should be more than 80%).

When to Use This Skill

Trigger this skill when:

  • Workflows added/modified — After editing .claude/workflows.json
  • Development rules changed — After editing .claude/docs/development-rules.md
  • Project-reference docs updated — After modifying files in docs/project-reference/
  • Registry entries changed — After editing docs/copilot-registry.json
  • Regular maintenance — Quarterly sync to ensure Copilot parity
  • Copilot setup — First-time Copilot instructions creation

Workflow

Phase 1: Script Generation (Automated)

node .claude/scripts/sync-copilot-workflows.cjs

This generates:

  • .github/copilot-instructions.md — project-specific with registry summaries
  • .github/instructions/common-protocol.instructions.md — generic protocols
  • .github/instructions/{group}.instructions.md — per-group instruction files
  • Removes old .github/common.copilot-instructions.md if it exists

Phase 2: AI Enrichment (This Skill)

After the script runs, the AI MUST ATTENTION enrich the generated instruction files:

  1. For each per-group instruction file in .github/instructions/:

- Read the corresponding docs/project-reference/*.md source file - Extract the ## section headings from the source file - Add a "Key Sections" list under each doc entry showing the headings - Keep it concise — headings only, no content duplication

  1. For .github/copilot-instructions.md:

- Verify the TL;DR golden rules in docs/copilot-registry.json → projectInstructions.goldenRules still match CLAUDE.md - Check if any new project-reference files exist but are missing from docs/copilot-registry.json - If missing entries found, add them to the registry and re-run the script

  1. For docs/copilot-registry.json:

- Verify each summary field accurately describes the current file content - Update stale summaries based on actual file content - Add entries for any new docs/project-reference/*.md files

Phase 3: Verification

Check that:

  • .github/copilot-instructions.md contains project-specific content
  • .github/instructions/common-protocol.instructions.md contains protocols + workflow catalog
  • Per-group instruction files contain READ prompts
  • No old common.copilot-instructions.md file remains
  • Workflow count matches workflows.json
  • All project-reference files are represented in the registry

AI Enrichment Protocol

When enriching per-group instruction files, follow this pattern for each doc entry:

## [Doc Title](relative/path)

**Summary:** One-line summary from registry.

**Key Sections:**

- Section 1 Name
- Section 2 Name
- Section 3 Name
- ...

> **READ** `docs/project-reference/filename.md` when: trigger description

Rules:

  • Extract ONLY ## level headings from the source file (not ### or deeper)
  • Do NOT copy content — just list heading names
  • Keep the READ prompt from the registry whenToRead field
  • If a file is very large (>30KB), note the file size: (~59KB - read relevant sections)

Output Files

FileTypeContent
.github/copilot-instructions.mdProject-specificTL;DR + project-reference index + READ prompts
.github/instructions/common-protocol.instructions.mdGeneric (applyTo: **/*)Prompt protocol + workflow catalog + dev rules
.github/instructions/backend-csharp.instructions.mdBackend (applyTo: **/*.cs)Backend doc summaries + READ prompts
.github/instructions/frontend-angular.instructions.mdFrontend (applyTo: **/*.ts,**/*.html)Frontend doc summaries + READ prompts
.github/instructions/styling-scss.instructions.mdStyling (applyTo: **/*.scss,**/*.css)Styling doc summaries + READ prompts
.github/instructions/testing.instructions.mdTesting (applyTo: **/*Test*/**,...)Testing doc summaries + READ prompts
.github/instructions/project-reference.instructions.mdCross-cutting (applyTo: **/*)General project doc summaries + READ prompts

Copilot Limitations

Copilot can't enforce protocols like Claude Code hooks:

  • No blocking operations (edit-enforcement, search-before-code)
  • Relies on LLM instruction-following (not guaranteed)
  • Protocols are advisory, not enforced
  • No runtime context injection — all context must be in instruction files

Benefits:

  • Consistent guidance across AI tools
  • Same workflow detection for Claude and Copilot users
  • READ prompts enable on-demand context loading
  • Automated sync reduces configuration drift

Troubleshooting

Issue: "workflows.json not found"

Solution: Ensure you're running from project root

Issue: Missing project-reference files in registry

Solution: Add entries to docs/copilot-registry.json, then re-run script

Issue: Stale summaries

Solution: Run this skill — AI will read files and update summaries


Related Skills

  • /ai-dev-tools-sync — Broader Claude/Copilot sync (skills, prompts, agents)
  • /sync-copilot-workflows — Workflow-only sync (subset of this skill)

References

  • Script: .claude/scripts/sync-copilot-workflows.cjs
  • Registry: docs/copilot-registry.json
  • Sources: .claude/workflows.json, .claude/docs/development-rules.md
  • Main output: .github/copilot-instructions.md
  • Instruction files: .github/instructions/*.instructions.md

Closing Reminders

  • IMPORTANT MUST ATTENTION break work into small todo tasks using TaskCreate BEFORE starting
  • IMPORTANT MUST ATTENTION search codebase for 3+ similar patterns before creating new code
  • IMPORTANT MUST ATTENTION cite file:line evidence for every claim (confidence >80% to act)
  • IMPORTANT MUST ATTENTION add a final review todo task to verify work quality

平台分布

Codex

33.3%
按下载量换算31

Claude

31.83%
按下载量换算29

Cursor

19.14%
按下载量换算18

Gemini CLI

9.83%
按下载量换算9

安全审计

Gen Agent Trust Hub

通过

Socket

通过

Snyk

通过

权限和风险

只读

该 Skill 主要提供规则、说明或参考内容,本身偏只读;真正读写文件、联网或执行命令仍取决于宿主 Agent 的任务。

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。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