If you build with an AI agent, you already know the feeling: the agent can write your code, but it cannot see your customers. Your inbox lives in one tab, your work lives in another, and you copy-paste between them all day.
muro closes that gap. It ships a hosted MCP server, so the agent you already use, Claude Code, Cursor, Claude Desktop, or your own, can read and act on your support inbox directly.
What the agent can do
- →List and read conversations and contacts.
- →Draft and send replies to visitors.
- →Tag, assign, snooze, and resolve threads.
- →Read and update your sites.
Connect it in one paste
In muro, open the AI Agent tab and create an MCP connection. You get a ready-to-paste config for your client, plus a one-line command for Claude Code:
bashclaude mcp add --transport http muro https://muro.chat/api/mcp \
--header "Authorization: Bearer mr_live_…"Why this is the unlock
When a customer reports a bug, your agent can read the conversation, look at the code they are complaining about, check the fix, and answer them, without leaving the editor. That is support that scales with one person, not with headcount.
API and MCP are on every plan, not gated behind an enterprise tier. Read the docs or start a trial.