$ loading_
Reliably relay jobs between agents and wake Claude when work completes.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mcp-agent-relay" yet — see the docs or source repo.
Use mcp-agent-relay to dispatch the task 'summarize this week's API changes and generate release notes' to another agent. When finished, wake this Claude session through the specified channel and return job status, result summary, and retry details if it fails.
Creates a relay job, tracks execution status, and returns the completion summary back to the session.
Send a code review task to a review agent through mcp-agent-relay: inspect this PR for possible bugs, performance issues, and test gaps. When the job is done, automatically wake my current Claude session and provide findings plus suggested fixes.
Receives confirmation of the async review job and a structured review report on completion.
Use mcp-agent-relay to submit a long-running task so a background agent can generate a project dependency inventory and risk analysis. Do not block this session; when finished, wake me through the channel and include a result link or summary.
The current session stays unblocked and later receives a wake-up notification with the analysis results.
Stream real-time agent plans, progress, tool calls, and results locally.
Connect Claude Desktop and Claude Code for autonomous message, file, and code exchange.
Access and orchestrate a unified catalog of API tools via MCP.
Run a persistent local MCP relay with caching and hot server refresh.
Control Claude Code sessions in tmux with fine-grained orchestration and efficient extraction.
Relay asynchronous Q&A between separate Claude Code instances without shared context.