经 AI Skill Hub 精选评估,子代理技能 获评「强烈推荐」。这款Agent工作流在功能完整性、社区活跃度和易用性方面表现出色,AI 评分 8.0 分,适合有一定技术背景的用户使用。
子代理编排,支持Codex、Claude等
子代理技能 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
子代理编排,支持Codex、Claude等
子代理技能 是一套完整的 AI Agent 自动化工作流方案。通过可视化的节点编排,将复杂的多步骤任务拆解为清晰的自动化流程,实现全程无人值守的智能处理。支持与数百种外部服务和 API 无缝集成,适合构建数据处理管线、业务自动化和 AI 辅助决策系统。
# 方式一:pip 安装(推荐)
pip install sub-agents-skills
# 方式二:虚拟环境安装(推荐生产环境)
python -m venv .venv
source .venv/bin/activate # Windows: .venv\Scripts\activate
pip install sub-agents-skills
# 方式三:从源码安装(获取最新功能)
git clone https://github.com/shinpr/sub-agents-skills
cd sub-agents-skills
pip install -e .
# 验证安装
python -c "import sub_agents_skills; print('安装成功')"
# 命令行使用
sub-agents-skills --help
# 基本用法
sub-agents-skills input_file -o output_file
# Python 代码中调用
import sub_agents_skills
# 示例
result = sub_agents_skills.process("input")
print(result)
# sub-agents-skills 配置文件示例(config.yml) app: name: "sub-agents-skills" debug: false log_level: "INFO" # 运行时指定配置文件 sub-agents-skills --config config.yml # 或通过环境变量配置 export SUB_AGENTS_SKILLS_API_KEY="your-key" export SUB_AGENTS_SKILLS_OUTPUT_DIR="./output"
Orchestrate any LLM as a sub-agent from any AI coding tool.
Use Codex, Claude Code, Cursor CLI, and Gemini CLI as sub-agents within a single workflow — regardless of which tool you're running. Define task-specific agents once in markdown, and execute them on any backend.
Codex (plugin):
codex plugin marketplace add shinpr/sub-agents-skills
Then open the plugin picker, install Runner, and restart Codex:
/plugins
After restart, invoke the skill as $runner:sub-agents.
Claude Code (plugin):
/plugin marketplace add shinpr/sub-agents-skills
/plugin install runner@sub-agents-skills
/reload-plugins
Other clients (Cursor CLI, Gemini CLI, VS Code, etc.):
Use the install script to copy the skill into the client's skill path:
```bash
Requirements: Python 3.9+ and at least one supported backend installed.
To run an agent, describe the task in your prompt:
"Use the code-reviewer agent to check my UserService class"
"Use the test-writer agent to create unit tests for the auth module"
"Use the doc-writer agent to add JSDoc comments to all public methods"
The host tool invokes the agent and returns results.
Mixing backends in one project:
You can have agents that use different LLMs side by side:
.agents/
├── test-writer.md # run-agent: codex (fast generation)
├── code-reviewer.md # run-agent: claude (strong reasoning)
└── doc-writer.md # run-agent: gemini (large context window)
"Use the code-reviewer agent to find security issues, then use the test-writer agent to add tests for the fixes"
Tip: Always include what you want done in your request—not just which agent to use. Specific prompts get better results.
Each .md or .txt file in your .agents/ folder becomes an agent. The filename becomes the agent name (e.g., bug-investigator.md → "bug-investigator").
bug-investigator.md
| Field | Values | Description |
|---|---|---|
run-agent | codex, claude, cursor-agent, gemini, glm | Which CLI executes this agent |
permission | read-only, safe-edit (default), yolo | Approval/sandbox level the sub-agent runs with |
If run-agent is not specified, the skill auto-detects the caller environment or defaults to codex.
Permission levels:
read-only — investigation/review only, no edits or shell writes (codex -s read-only / claude --permission-mode plan / gemini --approval-mode plan / cursor --mode plan)safe-edit — auto-approve edits inside the workspace, suppress prompts (default; codex -s workspace-write + approval_policy=never / claude --permission-mode acceptEdits / gemini --approval-mode auto_edit / cursor --trust)yolo — bypass all approvals and sandboxing. Use with care.Sub-agents have no stdin, so any approval prompt would deadlock the run. The default safe-edit keeps writes confined to the workspace while suppressing prompts.
--cli argument (explicit override)run-agent frontmattercodex--cli always overrides the agent definition's run-agent.
Install the required CLI: - Codex: npm install -g @openai/codex - Claude Code: curl -fsSL https://claude.ai/install.sh | bash - Cursor CLI: curl https://cursor.com/install -fsS | bash - Gemini CLI: npm install -g @google/gemini-cli
curl -fsSL https://raw.githubusercontent.com/shinpr/sub-agents-skills/main/install.sh | bash -s -- --target .github/skills
Or clone manually:
bash git clone https://github.com/shinpr/sub-agents-skills.git cd sub-agents-skills ./install.sh --target <client-skill-path> ```
高质量的AI工作流项目,支持多种LLM子代理
AI Skill Hub 为第三方内容聚合平台,本页面信息基于公开数据整理,不对工具功能和质量作任何法律背书。
建议在沙箱或测试环境中充分验证后,再部署至生产环境,并做好必要的安全评估。
✅ MIT 协议 — 最宽松的开源协议之一,可自由商用、修改、分发,仅需保留版权声明。
AI Skill Hub 点评:子代理技能 的核心功能完整,质量优秀。对于自动化工程师和运维人员来说,这是一个值得纳入个人工具库的选择。建议先在非生产环境试用,再逐步推广。
| 原始名称 | sub-agents-skills |
| Topics | ai-agentscodexclaudepython |
| GitHub | https://github.com/shinpr/sub-agents-skills |
| License | MIT |
| 语言 | Python |
收录时间:2026-07-07 · 更新时间:2026-07-07 · License:MIT · AI Skill Hub 不对第三方内容的准确性作法律背书。
选择 Agent 类型,复制安装指令后粘贴到对应客户端