$ loading_
Access and orchestrate a unified catalog of API tools via MCP.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "Relay" yet — see the docs or source repo.
Using Relay, list the currently available tool catalog and group it by OpenAPI, GraphQL, MCP, and Google Discovery. For each group, provide the number of callable tools and a few example names.
A protocol-based summary of available tools with counts and representative examples.
I want to build an AI assistant that looks up user information, creates support tickets, and sends notifications. Through Relay, identify the most suitable tools or APIs and explain their purposes and calling order.
Recommended tools, an execution sequence, and an explanation of which API to call at each step.
Based on the tools accessible through Relay, design an AI agent workflow for collecting sales data, summarizing it into a report, and posting it to a team channel. Include inputs, outputs, and retry strategies.
An end-to-end agent workflow plan including tool selection, step dependencies, and error-handling guidance.
Use one MCP server for files, web search, screenshots, and planning workflows.
Reliably relay jobs between agents and wake Claude when work completes.
Connect local Codex agents to AgentRelay for orchestration and task management.
Self-host shared memory, structured relays, and traceable evidence for AI runtimes.
Run a persistent local MCP relay with caching and hot server refresh.
Orchestrate AI team tasks with free models and Claude via MCP.