PigeonRelay
ai.pigeonrelayv1.0.0Updated Sep 29, 2026
Delegate tasks between AI agents on different platforms and collect the results.
Overview
AI-generated overview
PigeonRelay lets an assistant hand off tasks to AI agents on other platforms and gather their results.
- What it does
- PigeonRelay is a remote MCP endpoint for delegating work between AI agents that run on different platforms. According to its registry description, it lets an assistant pass tasks to other agents and then collect the results back. No tool list, README, or configuration details are published, so the exact operations and result formats are not documented here.
- When to use it
- Consider it when a workflow spans several agent platforms and you want one assistant to dispatch work to another and retrieve the outcome. It is less useful if all your agents already share a single platform or if you need documented, auditable tool behavior before adopting it.
- Requirements
- It is a remote streamable HTTP endpoint at mcp.pigeonrelay.ai; nothing needs to be installed locally. The manifest declares no authentication, environment variables, or headers, and no package is published. Network access to the endpoint is required.
Before you install
The manifest declares no authentication, so it is unclear how access is controlled or who can reach the endpoint. Tasks and their results are relayed through a third-party service, meaning content you delegate leaves your environment. No tool list is published, so what the server can read, send, or change is undocumented; verify this before relying on it.
Installation
In SourceWeft
- Open PigeonRelay in the dashboard and add it to a workspace.
- Enable the server for the chats that should use its tools.
Web executable via Streamable HTTP. Remote servers run from the web runtime once configured in a workspace.
Other MCP clients
Add this to your client's mcpServers config.
{
"mcpServers": {
"pigeonrelay": {
"type": "http",
"url": "https://mcp.pigeonrelay.ai/mcp"
}
}
}Tools
0Tool metadata has not been indexed yet.
Version history
1- v1.0.0LatestSep 29, 2026
