agentchan MCP server
Commit Oct 1, 2026Enables communication between different coding agents like Claude Code, Codex, and Grok using channel-based messaging.
Also known as damejeras/agentchan
- Stars
- 0
- Tools
- —
- Downloads, 30 days
- —
- Latest version
- 0.1.4
- Last commit
- Oct 1, 2026
- License
- MIT
What is agentchan MCP?
agentchan enables communication between different coding agents like Claude Code, Codex, and Grok using channel-based messaging.
Summary generated from the project's README
About agentchan MCP server
Facilitating direct communication between different coding agents like Claude Code, Codex, and Grok is the primary function of agentchan. It allows developers to create multi-agent workflows by treating individual agent sessions as message channels. When you start the server, it automatically provides tools for every agent that is already installed and signed in on your local system.
Users interact with the server through specific tools such as claude, codex, or grok to send messages and initiate sessions. The receive tool allows a client to wait up to 50 seconds for a reply on specified channels, while the close tool terminates an agent and its associated channel. Because each message executes commands like claude -p or grok --prompt-file in the server's working directory, the agents operate using their existing configurations and plan limits. Note that agents run with permission checks disabled, granting them full read, write, and execution access within the working directory.
What you can do with it
- 1Multi-agent collaborationA developer can coordinate a workflow where one agent sends a task to another agent via a specific channel.
- 2Asynchronous agent responsesYou can instruct an AI client to wait for a response from a specific agent channel for a set duration.
Written from the project's README. Every use case is checked against a quote from it.
Automated checks
Ran Nov 15, 2026Full report →Not verified yet. A server needs a score of 80, no serious finding, and enough data for us to check. How checks work · What to fix for more points
- Source repository linked and reachabledamejeras/agentchan is public and reachablegithub.com/damejeras/agentchan ↗Pass
- Package names this repository as its sourceNot checked
- Build provenance (Sigstore attestation)Not checked
- Open source license (OSI)MITspdx.org/licenses/MIT.html ↗Pass
- Commit or release in the last yearLast commit or release 2026-10-01github.com/damejeras/agentchan ↗Pass
- OpenSSF ScorecardNot checked
- Package name not close to a popular packageNot checked
- No critical advisory in the package or its dependenciesNot checked
- No hidden instructions to the model0 tool descriptions and the README read; no override phrases, credential paths or hidden charactersPass
- Destructive tools are labelledNot checked
- Remote endpoint answers the MCP handshakeNot checked
README
View on GitHubagentchan
An MCP server that lets one coding agent talk to another: Claude Code, Codex or Grok. It gives one tool for each agent that is installed and signed in, plus two tools that work like Go channels:
claude(message, channel?),codex(…),grok(…)send a message and return a channel id at once. Withoutchannel, the message starts a new session. Withchannel, it is a follow-up.receive(channels)waits for the first reply on any of the channels, up to 50 s.close(channel)stops the agent and ends the channel.
Each channel is one agent session. Each message runs claude -p, codex exec or grok --prompt-file
in the server's working directory, with permission checks off, and with the settings that agent already has.
Install
Linux and macOS, prebuilt:
curl --proto '=https' --tlsv1.2 -LsSf https://github.com/damejeras/agentchan/releases/latest/download/agentchan-installer.sh | sh
With npm, which downloads the same prebuilt binary:
npm install -g agentchan
From source:
cargo install agentchan
The server offers every agent that is installed and signed in when it starts.
# Claude Code
claude mcp add --scope user agentchan -- agentchan
# or, without installing it first
claude mcp add --scope user agentchan -- npx -y agentchan
# Codex: ~/.codex/config.toml
[mcp_servers.agentchan]
command = "agentchan"
Security
The called agent runs with permission checks off. It can read, write and run anything in the server's working directory without asking. Any agent that can call agentchan gets that access.
When a turn ends, and when a channel closes or the server stops, agentchan kills the agent's process group. A process that the agent moved to a different process group keeps running.
Test
cargo test # fake agents, no plan limits used
cargo test -- --ignored # the real CLIs that are signed in
Cost
Each message uses the called agent's own login and plan limits. Each message starts a new process that loads that agent's full configuration again.
Links
- MCP Registry name:
mcp-name: io.github.damejeras/agentchan
agentchan MCP questions
Which agents are supported by agentchan?
The server supports Claude Code, Codex, and Grok, provided they are installed and signed in.
Does using agentchan incur extra costs?
No, each message uses the existing login and plan limits of the specific agent being called.
What are the security implications of using this server?
The called agent runs with permission checks disabled, meaning it can read, write, and execute anything in the server's working directory.