Claude-to-IM
claude-to-im
A bridge between Claude Code or Codex and messengers: talk to the agent from Telegram, Discord, Feishu, QQ or WeChat
Install
npx skills add op7418/Claude-to-IM-skillAfter install tell Claude Code: /claude-to-im setup.
This is third-party code. Review the repository files before installing.
What it does
The skill runs a background Node.js daemon that links messenger bots to a Claude Code or Codex session. Chat messages go to the agent, and its replies, permission requests and previews come back to the messenger. Five platforms are supported: Telegram, Discord, Feishu, QQ and WeChat, in any combination. Every tool call needs explicit approval through buttons or text commands, and tokens are stored with mode 600 and redacted from logs. Setup runs through a wizard via the setup command.
Who it is for. For developers who want to task the agent and watch its replies from their phone.
Good fit when
- You want to run Claude Code tasks from your phone via Telegram or Discord
- You need to approve agent actions while away from the computer
- You work in Feishu and want to link it to the agent
Not a fit when
- You need a plain bot or webhook without a coding agent
- You are not ready to run a background daemon and hold bot tokens on your machine
Example request
claude-to-im setup, I want to connect a Telegram botLimitations
Requires Node.js 20 and an installed, authenticated Claude Code or Codex. You need bot tokens for the relevant messengers, and QQ and WeChat have reduced feature sets. The daemon runs locally and opens no inbound network ports.
How to disable. Stop the daemon with /claude-to-im stop and remove the skill folder from the agent's skills directory, optionally deleting the ~/.claude-to-im directory.
Security check
- Runs a background daemon and stores bot tokens
- Exposes agent control from an external chat, limited by allow lists
README in short
The README explains that the skill was extracted from the CodePilot desktop app to give a lightweight CLI-only option. The daemon connects bots to the agent, supports streaming preview in Telegram and Discord, and keeps sessions across restarts. Each platform has a step-by-step wizard that points to where tokens come from and which permissions to enable. The setup, start, stop, status, logs, reconfigure and doctor commands manage the bridge from inside the agent. MIT licensed.
SKILL.md
--- name: claude-to-im description: | Bridge THIS Claude Code or Codex session to Telegram, Discord, Feishu/Lark, QQ, or WeChat so the user can chat with Claude from their phone. Use for: setting up, starting, stopping, or diagnosing the claude-to-im bridge daemon; forwarding Claude replies to a messaging app; any phrase like "claude-to-im", "bridge". Subcommands: setup, start, stop, status, logs, reconfigure, doctor. --- # Claude-to-IM Bridge Skill You are managing the Claude-to-IM bridge. User data is stored at `~/.claude-to-im/`.
FAQ
Who can message the bot?
Access is restricted by allowed user, channel and server lists in the config.
Do I need OpenAI keys?
No, if you use Claude Code. For Codex you need its own auth, and an OpenAI key is optional.
Related
A CLI for every Google Workspace API with JSON output and agent skills: Drive, Gmail, Calendar, Sheets and more
Local search over Markdown notes, docs and meeting transcripts: keywords, semantic search and reranking, with an MCP server
A task manager for AI-driven development: breaks a PRD into dependent tasks and guides the agent through them via MCP or CLI
Anthropic's official plugins for knowledge work roles: sales, support, product, marketing, legal, finance, data and enterprise search