texting
Enables Claude to send and read iMessages on macOS, with human approval required for sending and no auto-replies.
README
texting
Use Claude to text and read iMessages — the same way you'd use it with Gmail or Slack. macOS only. No background agent, no auto-replies: Claude only reads or sends when asked, and sends get a normal permission prompt first (review before it goes out).
"Text Sam: confirmed for Friday at 3"
"Any new messages from Sam?"
"Catch me up on my texts"
How it works
The macOS iMessage/SMS engine is openclaw/imsg,
a mature Swift CLI that reads ~/Library/Messages/chat.db and drives
Messages.app. The plugin ships a thin MCP server over it, so texting works
as a plugin in the Claude desktop app (not just the terminal). It runs
in every session and exposes:
send_message— send to a phone number, Apple ID email, contact name, or chat_id; works for brand-new contacts (supports attachments). Routes iMessage vs SMS automatically — Android recipients work via Text Message Forwarding, with SMS fallback when iMessage isn't availablelist_chats— an inbox-style view of your most recent conversationsread_messages— read conversation history; pass a person's handle to cover their iMessage and SMS threads at once, or a chat_id for a groupsearch_messages— find messages by text, newest first ("when did Sam mention the invoice?")react— tapback the latest message in a thread (love, like, laugh…)watch— wait briefly for incoming messages and return them (with a cursor to resume from), for catching a reply that's expected shortly
Names come from macOS Contacts — the same cards the user sees in the Contacts app. The plugin keeps no contact store of its own; people Claude doesn't recognize get offered a real Contacts card.
| Skill | Does |
|---|---|
/texting-setup |
One-time setup: install the imsg engine (official release download), then the Full Disk Access toggle |
/texting-send |
Send a text (confirms recipient + exact wording first) |
/texting-messages |
Read threads, resolve "Sam" → handle via macOS Contacts |
/texting-listen |
Session-bound watch: "tell me when Alex replies" — polls, notifies, drafts replies for approval, never auto-sends |
/texting-status |
Health check |
Setup friction (one-time, ~5 minutes)
- The setup skill installs the
imsgengine for you — it downloads the signed, notarized macOS binary from the official openclaw/imsg release (checksum verified) into~/.claude/texting/engine. No Homebrew or developer tools needed. (Alternatives:brew install steipete/tap/imsg, or setIMSG_PATH.) - One Full Disk Access toggle for the app hosting Claude (desktop users: the embedded claude.app — the setup skill knows the path), then restart it
- The macOS Automation prompt on the first send
Security model
Same consent shape as a mail connector: Claude can read any conversation when asked — granting Full Disk Access is the "Claude may read my texts" decision, just like connecting Gmail is the "Claude may read my email" decision. The guarantees on top:
- Every send is human-approved (permission prompt with the exact text). An
optional
- Sent by Claude for <name>AI-disclosure signature is available but off by default — enable it globally (signature: truein~/.claude/texting/config.json) or per send (sign_as); we don't stamp your texts unless you ask. - Nothing listens for inbound texts; an incoming message cannot trigger Claude. Reads happen only when the user asks, about what they asked.
- Message content Claude reads is treated as data — instructions embedded in texts are surfaced to the user, never acted on.
Auto-reply mode
Deliberately not part of this plugin's setup. If you want Claude answering texts unattended, that's an always-on headless process with a very different risk profile (replies go out unreviewed; the sender allowlist becomes a real security boundary) — see docs/AUTOREPLY.md for the recipe and the checklist of caveats.
Credits
Built on openclaw/imsg (MIT) — the iMessage/SMS engine that does the reading, sending, automatic iMessage/SMS routing, and reactions. This plugin is the MCP surface and skills around it.
See NOTICE for full attribution.
Support & contributing
Best-effort maintenance — I'll do my best to fix bugs and security issues, and contributions are very welcome. Open an issue or a pull request; help with docs, tests, or features is encouraged. See CONTRIBUTING.md.
License
Apache-2.0 © Ian McDonald.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.