myagents-cli

Operate MyAgents product capabilities through the built-in myagents CLI.

863|101|Updated Jan 23, 2026
One-click install
npx skills add https://github.com/hAcKlyc/MyAgents --skill myagents-cli
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: myagents-cli
Source: https://github.com/hAcKlyc/MyAgents/tree/main/bundled-skills/myagents-cli
Command: npx skills add https://github.com/hAcKlyc/MyAgents --skill myagents-cli

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Users of the MyAgents desktop agent platform often need to configure scheduled tasks, MCP tools, model providers, IM bot channels, plugins, skills, and cloud spaces, but doing so through the GUI is slow and interrupts the conversation flow. This Skill lets the AI execute those product operations directly via the built-in myagents CLI, so requests like "remind me every day at 9" or "connect a Notion MCP tool" are completed in-session instead of requiring manual settings navigation.

Core Features & Use Cases

  • Task & Automation Management: Create, schedule, pause, and rerun one-time or recurring tasks, including command-detector-based conditional triggers and Goal mode for continuous execution.
  • Platform Configuration: Manage MCP servers (add, enable, OAuth, env vars), model providers (API keys, defaults, verification), agents, channels (Telegram, DingTalk, OpenClaw), plugins, and skills installation from GitHub or local sources.
  • Cloud Space & Collaboration: Create and claim Cloud Space issues, post comments with attachments, download files, and coordinate across sessions with session start/send/watch.
  • Use Case: A user says "set up DeepSeek as my default model and schedule a daily code review at 9am." The Skill runs myagents model set-key deepseek <key>, model verify, model set-default, then creates and starts a scheduled task — all without the user opening Settings.

Quick Start

Ask the AI to configure a MyAgents capability in natural language, for example: "Connect the fetch MCP tool and schedule a task that checks my server status every morning at 8am."

Frequently Asked Questions about myagents-cli

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I schedule a recurring task in MyAgents?

Use the myagents task commands to create a scheduled task with an interval, cron expression, or dispatch time, then run `myagents task start <taskId>` to activate it. The myagents-task-automation skill handles choosing between always-active schedules and command detectors.

How do I connect an MCP tool to MyAgents?

Run `myagents mcp add` with the server ID, transport type (stdio, sse, or http), and command or URL, then enable it with `mcp enable --scope both` and verify with `mcp test`. MCP tools become available in the next conversation turn, not the current one.

How do I configure a model provider like DeepSeek in MyAgents?

For built-in providers, run `myagents model set-key deepseek <apiKey>` followed by `model verify` to test connectivity. For new vendors, use `model add` with the base URL and model list, preferring the Anthropic protocol over OpenAI-compatible endpoints when available.

Why does the myagents CLI return command not found or ECONNREFUSED?

A command-not-found error means the CLI has not been synced to ~/.myagents/bin yet, so restart the application once. ECONNREFUSED means the Sidecar Admin API is not running, so check that the MyAgents app is active.

Can I install skills from GitHub or local directories?

Yes, `myagents skill add` accepts GitHub owner/repo shorthand, full URLs, repository subpaths, zip links, and local directories with explicit ./ or ../ prefixes. It does not support tar.gz archives, GitLab, private repositories, or git SSH sources.

What is the difference between MyAgents plugins and Claude plugins?

The `plugin` command manages OpenClaw IM channel plugins installed from npm, while `cc-plugin` manages Anthropic Claude Plugin protocol packages containing skills, agents, MCP servers, and hooks. The two systems are independent and do not conflict.