Agent skill · Frontend

myagents-cli

你正在 MyAgents 这款 AI 产品里运行——MyAgents 自带一套"产品能力"（定时任务、任务中心、想法收集、MCP 工具接入、 模型 Provider、IM Bot 渠道、社区插件、Skills 安装、MyAgents Cloud Space、Generative UI Widget、Goal 目标模式等），全部通过内置 `myagents` CLI 暴露给你。 当用户的需求**落在 MyAgents 产品能力的射程内**，就加载并使用这个 skill，用 CLI 主动帮用户把事情做掉， 而不是让用户去 GUI 点击。 典型触发场景：用户说"每天 X 点帮我 Y / 等 X 发生后继续 / 持续盯着，命中才处理"（→ myagents-task-automation）、"记一下这个想法"（→ thought）、"派发成任务"（→ task）、 "接个 X 工具进来"（→ mcp）、"配 X 模型/Provider"（→ model）、"在飞书/钉钉/Telegram 里跟我聊"（→ agent channel）、 "装个 X 插件 / 装个 X skill"（→ plugin / skill）、"处理 Space Issue / 下载附件 / 回复 Issue"（→ space）、 "把图发到 IM 里"（→ im send-media）、"用已配置的读图模型理解图片

hAcKlycgithub.com/hAcKlycGitHub ↗
claude-codeAGPL-3.0
Install
npx skills add hAcKlyc/MyAgents --skill myagents-cli --agent claude-code

Same command for any agent — swap --agent for codex, cursor, copilot.

Facts
Files in the skill folder: 1
SKILL.md size: 31 KB
Bundled scripts: none
Declared author: MyAgents
Path: bundled-skills/myagents-cli/SKILL.md
Open the folder on GitHub →
Where it comes from
Stars: 795
Language: TypeScript

Weekly change comes from our own snapshots, not the repository page — it measures attention, not adoption.

Review
written from the skill's own SKILL.md · Aug 5, 2026

What it does

The skill exposes the MyAgents product capabilities through a CLI entry point. It instructs the agent to use the built-in myagents CLI to perform tasks that fall within the product’s automation, scheduling, task management, model/provider configuration, channels, widgets, and goal-oriented executions. It covers commands for MCP tools, models, agents and channels, vision image understanding, skills management, tasks, and goal mode. It also defines when to use the skill based on user intent mapping to these capabilities, and it emphasizes using the CLI for actions instead of GUI when applicable.

How it works

  • The agent is instructed to run commands via the CLI path: myagents <command>, which communicates through a Sidecar Admin API port defined by MYAGENTS_PORT.
  • It provides guidance on discovery and safe usage: use --help for unfamiliar groups, use --dry-run only for leaf commands that declare support, and use --json for machine-readable outputs.
  • For each capability (MCP, tool registry, vision, model, agent/channel, runtime, skills, task automation, goal), concrete myagents commands are listed with exact syntax and examples.
  • The skill defines when to invoke specific commands: for example, when user wants to connect a tool, configure a provider, chat via channels, install a skill, inspect vision, or run automated goal/task workflows.
  • It enforces safety rules such as API key handling and not executing user cancellation via commands unless the UI instructs.

When to use it

  • Triggered when user requests automation within the MyAgents capabilities, such as setting up a tool, configuring a provider, starting a task, or enabling a goal loop.
  • Use the relevant command group (mcp, model, agent, vision, tool, task, goal, widget/readme) according to the user’s mapped intent.
  • If the user asks to see status, list items, or modify configurations, follow the corresponding myagents command sequences provided.

What it can touch

  • Commands touch: MCP tools (myagents mcp ...), CLI tools registry (myagents tool ...), official vision tool (myagents vision ...), model providers (myagents model ...), Agent/Channel configuration (myagents agent ...), runtime queries (myagents runtime ...), Skills (myagents skill ...), task automation (myagents task ...), and Goal mode (myagents goal ...).

Caveats

  • The skill relies on the presence and accessibility of the CLI at ~/.myagents/bin/myagents and the Sidecar API port MYAGENTS_PORT. If command not found or ECONNREFUSED, user actions are required to fix the runtime state.
  • Some advanced features require specific preconditions (e.g., certain leaf commands or experimental features behind feature gates). The guidance emphasizes using help outputs and exact leaf commands for safe previews.
From the SKILL.md

# myagents-cli — MyAgents 产品能力的 CLI 入口 你正运行在 MyAgents 产品内。MyAgents 不只是一个 chat UI，它是一套带状态的 Agent 平台：Goal 目标模式、定时任务、任务中心、IM Bot、MCP、Provider、插件、Skill、Cloud Space、Widget——这些都是产品能力，由内置 `myagents` CLI 一站暴露给你。 **这个 skill 不只是"管理工具"，它是 MyAgents 产品能力的执行入口**。用户表达的需求只要能映射到产品能力，就该用 CLI 主动帮用户做掉，而不是给用户一堆操作步骤让他自己去 Settings 点。这份文档列出全部能力以及"什么时候应该用哪条命令"。 ## 前置：CLI 是否可用 CLI 通过 `~/.myagents/bin/myagents` 暴露，你的 SDK 子进程 PATH 已注入这个目录，直接 `myagents <command>` 就能跑。它通过 HTTP 走 Sidecar Admin API（端口由环境变量 `MYAGENTS_PORT` 注入）。 - 遇到 `command not found`：让用户重启一次应用触发 CLI 同步 - 遇到 `ECONNREFUSED`：Sidecar 没起来，让用户检查应用是否在运行 ## 使用模式 1. **探索先行**：不熟的命令组用 `myagents <group> --help`；不知道某个 runtime 支持什么 model/permissionMode 用 `myagents runtime describe <runtime>`，**不要靠猜** 2. **按 leaf 契约预览**：只有精确 leaf help 明确声明支持的命令才使用 `--dry-run`；不支持的 mutation 会 fail closed，不能声称已预览 3. **机器可读**：加 `--json` 解析结构化输出 4. **失败即恢复**：CLI 失败响应会带 `→ Run: <cmd>` 恢复提示，照着跑就行 ## 安全规范 - **改配置前先读精确 leaf help**——该 leaf 明确支持 `--dry-run` 时先预览；未声明支持时不要假装存在 preview - **API Key**：用户在对话里明确给了你才写入；没给就引导他去 **设置 → 对应页面** 填，不要追问 - **删除前确认**：用户说"删了吧"也要回读"我要删的是 X，确认吗" ## 生效时机 - **MCP 工具变更**（增删改 / 启禁用 / 环境变量 / OAuth）：磁盘立即写入，但工具在**下一轮对话**才能调用——MCP server 在 session 创建时绑定。当前轮配完后告诉用

What's inside
Steps it walks through
  1. 前置：CLI 是否可用
  2. 使用模式
  3. 安全规范
  4. 生效时机
  5. 命令速查 + 何时使用
  6. MCP 工具（mcp）
  7. CLI 工具注册表（tool） — PRD 0.2.36，实验室开启后可用
  8. 官方图片理解工具（vision）
  9. 模型 Provider（model）
  10. Agent + Channel（agent）
  11. Agent Runtime 发现（runtime）
  12. Skills（skill）
  13. 定时与未来自动化 Task
  14. Goal 目标模式（goal）
Commands it runs
myagents mcp list                                       # 看用户配了哪些 MCP
myagents mcp show <id>                                  # 看某个 MCP 的完整配置（command/args/env/headers）
myagents mcp add --id <id> --type <stdio|sse|http> ...  # 新增
myagents mcp remove <id>                                # 删除
myagents mcp enable <id> --scope <user|project|both>    # 启用
myagents mcp disable <id> --scope <user|project|both>   # 禁用
myagents mcp test <id>                                  # 实际握手测试连通性
myagents mcp env <id> set KEY=val [KEY2=val2 ...]       # 设环境变量（覆盖）
myagents mcp env <id> get [KEY ...]                     # 读环境变量
myagents mcp env <id> delete KEY [KEY2 ...]             # 删环境变量
More from MyAgents
All skills →
download-anythingFind and download virtually any digital resource from the internet — ebooks, academic papers…Media & VideosupportMyAgents 本地问题诊断、恢复与反馈升级流程。用户描述报错、崩溃、无响应、配置后仍不可用、状态或结果不符合预期、…Workflow & Productivitymyagents-memory-molt仅当系统或用户明确指定完整名称 `myagents-memory-molt` 时使用； 不要根据任务语义或相似表述自行触发。AI & Agentsagent-browserBrowser automation CLI for AI agents. Use when the user needs to interact with websites, including…Testing & QAdocxComprehensive document creation, editing, and analysis with support for tracked changes, comments…Content & Marketinghatch-petCreate, repair, validate, visually QA, and package MyAgents/Codex-compatible animated pets and pet…Testing & QA
About this skill
What does the myagents-cli skill do?

你正在 MyAgents 这款 AI 产品里运行——MyAgents 自带一套"产品能力"（定时任务、任务中心、想法收集、MCP 工具接入、 模型 Provider、IM Bot 渠道、社区插件、Skills 安装、MyAgents Cloud Space、Generative UI Widget、Goal 目标模式等），全部通过内置 `myagents` CLI 暴露给你。 当用户的需求**落在 MyAgents 产品能力的射程内**，就加载并使用这个 skill，用 CLI 主动帮用户把事情做掉， 而不是让用户去 GUI 点击。 典型触发场景：用户说"每天 X 点帮我 Y / 等 X 发生后继续 / 持续盯着，命中才处理"（→ myagents-task-automation）、"记一下这个想法"（→ thought）、"派发成任务"（→ task）、 "接个 X 工具进来"（→ mcp）、"配 X 模型/Provider"（→ model）、"在飞书/钉钉/Telegram 里跟我聊"（→ agent channel）、 "装个 X 插件 / 装个 X skill"（→ plugin / skill）、"处理 Space Issue / 下载附件 / 回复 Issue"（→ space）、 "把图发到 IM 里"（→ im send-media）、"用已配置的读图模型理解图片

How do I install it?

Run `npx skills add hAcKlyc/MyAgents --skill myagents-cli --agent claude-code` — it drops the skill into your project so the agent can pick it up. Swap the --agent value for codex, cursor or copilot if you use one of those.

Where does this skill come from?

From hAcKlyc/MyAgents, a repository with 795 stars. We read it straight from the repository tree rather than a submitted listing, so what you see here is what is actually published.

Is a popular skill a good skill?

Not necessarily. Stars measure attention, not adoption — a repository can trend for a week and be abandoned. That is why we show the weekly change from our own snapshots next to the total, instead of a single flattering number.

Keep going
All skills →
The full catalog, by category and by tool.
Open Source Radar →
The repositories these skills come from, watched daily.
The people behind them →
Makers and builders across the ecosystem.