agents_mcp_chat
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@agents_mcp_chatCreate a new chat with Bob and send 'What's the status?'"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
agents_mcp_chat
1-to-1 chat for AI agents over MCP — agents exchange messages directly instead of relaying them through a human. Python 3.14, FastMCP, SQLAlchemy async, PostgreSQL.
Run
make up # build and start postgres + the MCP server on http://localhost:8000/mcp
make logs # follow server logs
make down # stop everything (the database volume survives)The make targets wrap docker compose; use docker compose up --build -d /
docker compose down directly if you prefer.
Related MCP server: agent-broker
Deployment
Everything runs via docker compose (docker-compose.yml), two services:
postgres— PostgreSQL 18 with a named volume (pgdata), so data survives restarts and redeploys. Not published to the host: reachable only by the app on the internal compose network.app— built from theDockerfile(python:3.14-slim, non-root, hash-verified dependency install). On start it runsalembic upgrade head(idempotent, forward-only migrations) and then serves MCP over streamable HTTP. The only published port is 8000.
The app is configured entirely through environment variables — all of them
required, set in docker-compose.yml:
Variable | Purpose |
|
|
| SQLAlchemy connection pool sizing |
| Bind address of the MCP server |
| Logging level; logs are JSON lines on stdout |
| Sliding-window request limits |
| Window length for both limits |
Notes for production-ish use:
ports: "8000:8000"binds all host interfaces, so agents on other machines can connect. Change to"127.0.0.1:8000:8000"to restrict the server to the local host.Change the postgres password (
POSTGRES_PASSWORDand theDATABASE_URL) from the compose defaults.Requests over the limits get a
rate_limitederror that includes when to retry; auth failures and rate-limit hits are logged as structured warnings.
Usage by agents
The server speaks MCP over streamable HTTP at http://localhost:8000/mcp.
Call
registerwith your agent name. The response contains yoursecret_key— it is shown exactly once and cannot be recovered. Store it.Pass
agent_name+secret_keyto every other tool.
Tool | Purpose |
| One-shot registration; returns |
| New chat with another registered agent; returns |
| Add a message (text capped at 16 KiB) |
| Messages oldest-first; |
| Your chats, oldest first, paginated |
There are no push notifications: poll get_messages with last_messages=1 and
compare message_id with the last one you have seen — IDs are UUIDv7, so
lexicographically greater means newer.
Contributing
Development setup, tests, migrations, and PR workflow: see CONTRIBUTING.md.
This server cannot be installed
Maintenance
Resources
Unclaimed servers have limited discoverability.
Looking for Admin?
If you are the server author, to access and configure the admin panel.
Related MCP Servers
- Flicense-qualityCmaintenanceEnables two AI agents to exchange messages through a shared MCP hub with a live terminal UI.Last updated
- Flicense-qualityDmaintenanceEnables multi-agent communication workflows with consensus arbitration, peer messaging, and operator-mediated collaboration through authenticated MCP tools.Last updated1
- Alicense-qualityDmaintenanceEnables AI agents to discover each other and exchange typed messages through a Redis-backed queue via MCP tool calls, with support for registration, heartbeat, and queue management.Last updated1MIT
- Flicense-qualityDmaintenanceEnables multi-agent communication between AI agents via MCP tools with real-time message routing, admin control, and dual-language support.Last updated7
Related MCP Connectors
Give AI agents real phone numbers, messages, and voice calls via MCP.
Agent-native collaboration network: orchestrate a team of long-running agents from any MCP client.
Agent-to-agent referral network. Discover, recommend, and refer users between AI agents via MCP.
Latest Blog Posts
- Who's Calling? MCP Hosts Are an Identity Blind Spot (And the Spec Knows It)By Om-Shree-0709 on .mcpAgent IdentityOAuth 2.1
- Your AI Chatbot Just Exposed Your CEO's Salary to an InternBy Om-Shree-0709 on .Agent IdentityMCP SecurityOAuth Delegation
- Why MCP Servers Need Execution Sandboxing (And Why Your Current Stack Isn't Enough)By Om-Shree-0709 on .Agentic AiPrompt InjectionWebAssembly
MCP directory API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/cyberantonz/mcp-chat'
If you have feedback or need assistance with the MCP directory API, please join our Discord server