$ loading_
Run a persistent local MCP relay with caching and hot server refresh.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "mcp-local-relay" yet — see the docs or source repo.
Use mcp-local-relay to set up a persistent local relay connecting filesystem, github, and browser MCP servers; enable tool discovery caching and keep-alive, and show how to add new servers through management tools.
A local relay setup plan describing connected MCP services, caching and keep-alive behavior, and how to add servers later.
My MCP agent can no longer reliably call the database server. Use mcp-local-relay to refresh that upstream connection without restarting other services, and check whether the tool discovery cache should be updated.
A refresh result for the target upstream server, confirming other services stay running and available tool state is updated.
Connect multiple MCP servers in my local development environment through mcp-local-relay, reduce repeated discovery overhead, and design a management flow that lets agents dynamically add or remove servers.
A unified access and management plan, including performance benefits from caching and an operational flow for dynamically adding or removing servers.
Reliably relay jobs between agents and wake Claude when work completes.
Aggregate remote MCP servers and route tools safely with prefixed names.
Manage local MCP agent connections, tool exposure, and upstream configs centrally.
Stream real-time agent plans, progress, tool calls, and results locally.
Connect multiple MCP servers through one gateway for unified tool access.
Run persistent stateful Python sessions with timeouts, isolation, and tool bridging.