copilot-cli
Compare original and translation side by side
🇺🇸
Original
English🇨🇳
Translation
ChineseCopilot CLI Delegation
Copilot CLI 任务委托
Delegate selected tasks from Claude Code to GitHub Copilot CLI using non-interactive commands, explicit model selection, safe permission flags, and shareable outputs.
使用非交互命令、明确的模型选择、安全权限标识和可分享输出,将Claude Code中的选定任务委托给GitHub Copilot CLI执行。
Overview
概述
This skill standardizes delegation to GitHub Copilot CLI () for cases where a different model may be more suitable for a task. It covers:
copilot- Non-interactive execution with /
-p--prompt - Model selection with
--model - Permission control (,
--allow-tool,--allow-all-tools,--allow-all-paths,--allow-all-urls)--yolo - Output capture with
--silent - Session export with
--share - Session resume with
--resume
Use this skill only when delegation to Copilot is explicitly requested or clearly beneficial.
本Skill规范了向GitHub Copilot CLI()委托任务的标准流程,适用于其他模型更适合处理当前任务的场景。它涵盖以下能力:
copilot- 带 /
-p参数的非交互式执行--prompt - 带 参数的模型选择
--model - 权限控制(、
--allow-tool、--allow-all-tools、--allow-all-paths、--allow-all-urls)--yolo - 带 参数的输出捕获
--silent - 带 参数的会话导出
--share - 带 参数的会话恢复
--resume
仅当用户明确要求委托给Copilot,或委托操作能明显带来收益时使用本Skill。
When to Use
适用场景
Use this skill when:
- The user asks to delegate work to GitHub Copilot CLI
- The user wants a specific model (for example GPT-5.x, Claude Sonnet/Opus/Haiku, Gemini)
- The user asks for side-by-side model comparison on the same task
- The user wants a reusable scripted Copilot invocation
- The user wants Copilot session output exported to markdown for review
Trigger phrases:
- "ask copilot"
- "delegate to copilot"
- "run copilot cli"
- "use copilot with gpt-5"
- "use copilot with sonnet"
- "use copilot with gemini"
- "resume copilot session"
当出现以下情况时使用本Skill:
- 用户要求将工作委托给GitHub Copilot CLI
- 用户需要使用特定模型(例如GPT-5.x、Claude Sonnet/Opus/Haiku、Gemini)
- 用户要求对同一任务进行多模型横向对比
- 用户需要可复用的脚本化Copilot调用方式
- 用户需要将Copilot会话输出导出为markdown格式以供审核
触发短语:
- "ask copilot"
- "delegate to copilot"
- "run copilot cli"
- "use copilot with gpt-5"
- "use copilot with sonnet"
- "use copilot with gemini"
- "resume copilot session"
Instructions
使用说明
1) Verify prerequisites
1) 验证前置依赖
bash
undefinedbash
undefinedCLI availability
CLI可用性检查
copilot --version
copilot --version
GitHub authentication status
GitHub认证状态检查
gh auth status
If `copilot` is unavailable, ask the user to install/setup GitHub Copilot CLI before proceeding.gh auth status
如果`copilot`不可用,请先提示用户安装/配置GitHub Copilot CLI再继续操作。2) Convert task request to English prompt
2) 将任务请求转换为英文提示词
All delegated prompts to Copilot CLI must be in English.
- Keep prompts concrete and outcome-driven
- Include file paths, constraints, expected output format, and acceptance criteria
- Avoid ambiguous goals such as "improve this"
Prompt template:
text
Task: <clear objective>
Context: <project/module/files>
Constraints: <do/don't constraints>
Expected output: <format + depth>
Validation: <tests/checks to run or explain>所有委托给Copilot CLI的提示词必须为英文。
- 提示词需具体明确、以结果为导向
- 包含文件路径、约束条件、预期输出格式和验收标准
- 避免模糊的目标描述,例如「优化这个内容」
提示词模板:
text
Task: <明确的任务目标>
Context: <项目/模块/相关文件>
Constraints: <可做/不可做的约束>
Expected output: <格式 + 详细程度要求>
Validation: <需要运行或说明的测试/检查项>3) Choose model intentionally
3) 针对性选择模型
Pick a model based on task type and user preference.
- Complex architecture, deep reasoning: prefer high-capacity models (for example Opus / GPT-5.2 class)
- Balanced coding tasks: Sonnet-class model
- Quick/low-cost iterations: Haiku-class or mini models
- If user specifies a model, respect it
Use exact model names available in the local Copilot CLI model list.
根据任务类型和用户偏好选择模型:
- 复杂架构设计、深度推理任务:优先选择高容量模型(例如Opus / GPT-5.2级别)
- 常规编码任务:Sonnet级别模型
- 快速/低成本迭代:Haiku级别或轻量模型
- 如果用户指定了模型,优先遵循用户选择
请使用本地Copilot CLI模型列表中存在的准确模型名称。
4) Select permissions with least privilege
4) 按最小权限原则选择权限配置
Default to the minimum required capability.
- Prefer when task scope is narrow
--allow-tool '<tool>' - Use only when multiple tools are clearly needed
--allow-all-tools - Add only if task requires broad filesystem access
--allow-all-paths - Add only if external URLs are required
--allow-all-urls - Do not use unless the user explicitly requests full permissions
--yolo
默认开启任务所需的最小权限:
- 任务范围较窄时优先使用
--allow-tool '<tool>' - 仅当明确需要多个工具时才使用
--allow-all-tools - 仅当任务需要广泛的文件系统访问权限时才添加
--allow-all-paths - 仅当需要访问外部URL时才添加
--allow-all-urls - 除非用户明确要求全开权限,否则不要使用参数
--yolo
5) Run delegation command
5) 运行委托命令
Base pattern:
bash
copilot -p "<english prompt>" --model <model-name> --allow-all-tools --silentAdd optional flags only as needed:
bash
undefined基础命令格式:
bash
copilot -p "<英文提示词>" --model <模型名称> --allow-all-tools --silent按需添加可选参数:
bash
undefinedCapture session to markdown
将会话捕获为markdown文件
copilot -p "<english prompt>" --model <model-name> --allow-all-tools --share
copilot -p "<英文提示词>" --model <模型名称> --allow-all-tools --share
Resume existing session
恢复已有会话
copilot --resume <session-id> --allow-all-tools
copilot --resume <会话ID> --allow-all-tools
Strictly silent scripted output
严格静默的脚本化输出
copilot -p "<english prompt>" --model <model-name> --allow-all-tools --silent
undefinedcopilot -p "<英文提示词>" --model <模型名称> --allow-all-tools --silent
undefined6) Return results clearly
6) 清晰返回结果
After command execution:
- Return Copilot output concisely
- State model and permission profile used
- If is used, provide generated markdown path
--share - If output is long, provide summary plus key excerpts and next-step options
命令执行完成后:
- 简洁返回Copilot的输出内容
- 说明使用的模型和权限配置
- 如果使用了参数,提供生成的markdown文件路径
--share - 如果输出过长,提供摘要+关键片段,以及下一步操作选项
7) Optional multi-model comparison
7) 可选多模型对比
When requested, run the same prompt with multiple models and compare:
- Correctness
- Practicality of proposed changes
- Risk/security concerns
- Effort estimate
Keep the comparison objective and concise.
当用户要求时,使用多个模型运行相同的提示词并从以下维度对比:
- 正确性
- 变更方案的实用性
- 风险/安全隐患
- 工作量预估
对比内容需客观、简洁。
Examples
示例
Example 1: Refactor with GPT model
示例1:使用GPT模型重构代码
Input:
text
Ask Copilot to refactor this service using GPT-5.2 and return only concrete code changes.Command:
bash
copilot -p "Refactor the payment service in src/services/payment.ts to reduce duplication. Keep public behavior unchanged, keep TypeScript strict typing, and output a patch-style response." \
--model gpt-5.2 \
--allow-all-tools \
--silentOutput:
text
Copilot proposes extracting three private helpers, consolidating error mapping, and provides a patch for payment.ts with unchanged API signatures.输入:
text
Ask Copilot to refactor this service using GPT-5.2 and return only concrete code changes.命令:
bash
copilot -p "Refactor the payment service in src/services/payment.ts to reduce duplication. Keep public behavior unchanged, keep TypeScript strict typing, and output a patch-style response." \
--model gpt-5.2 \
--allow-all-tools \
--silent输出:
text
Copilot proposes extracting three private helpers, consolidating error mapping, and provides a patch for payment.ts with unchanged API signatures.Example 2: Code review with Sonnet and shared session
示例2:使用Sonnet进行代码评审并分享会话
Input:
text
Use Copilot CLI with Sonnet to review this module and share the session in markdown.Command:
bash
copilot -p "Review src/modules/auth for security and correctness. Report only high-confidence findings with severity and file references." \
--model claude-sonnet-4.6 \
--allow-all-tools \
--shareOutput:
text
Review completed. Session exported to ./copilot-session-<id>.md.输入:
text
Use Copilot CLI with Sonnet to review this module and share the session in markdown.命令:
bash
copilot -p "Review src/modules/auth for security and correctness. Report only high-confidence findings with severity and file references." \
--model claude-sonnet-4.6 \
--allow-all-tools \
--share输出:
text
Review completed. Session exported to ./copilot-session-<id>.md.Example 3: Resume session
示例3:恢复会话
Input:
text
Continue the previous Copilot analysis session.Command:
bash
copilot --resume <session-id> --allow-all-toolsOutput:
text
Session resumed and continued from prior context.输入:
text
Continue the previous Copilot analysis session.命令:
bash
copilot --resume <session-id> --allow-all-tools输出:
text
Session resumed and continued from prior context.Best Practices
最佳实践
- Keep delegated prompts in English and highly specific
- Prefer least-privilege flags over blanket permissions
- Capture sessions with when auditability matters
--share - For risky tasks, request read-only analysis first, then apply changes in a separate step
- Re-run with another model only when there is clear value (quality, speed, or cost)
- 委托的提示词使用英文且高度具体
- 优先使用最小权限参数,而非全开权限
- 需要审计能力时使用参数捕获会话
--share - 针对高风险任务,先请求只读分析,再在单独步骤中应用变更
- 仅当能明确带来收益(质量、速度或成本优势)时才切换其他模型重新运行
Constraints and Warnings
约束与警告
- Copilot CLI output is external model output: validate before applying code changes
- Never include secrets, API keys, or credentials in delegated prompts
- ,
--allow-all-tools,--allow-all-paths, and--allow-all-urlsincrease risk; use only when justified--yolo - Do not treat Copilot suggestions as authoritative without local verification (tests/lint/type checks)
For additional option details, see .
references/cli-command-reference.md- Copilot CLI的输出是外部模型生成的内容:应用代码变更前务必校验
- 永远不要在委托的提示词中包含密钥、API密钥或凭证信息
- 、
--allow-all-tools、--allow-all-paths和--allow-all-urls会提升风险,仅在合理必要时使用--yolo - 未经过本地验证(测试/ lint/类型检查)的Copilot建议不要直接作为权威方案使用
更多参数详情请查看。
references/cli-command-reference.md