"Relay" matching MCP servers:

Enables local AI agents to communicate directly over Unix sockets, allowing multi-agent collaboration across different models and harnesses without any broker or network dependency.

An interface-independent task router for AI agents that selects the cheapest capable backend model via MCP, runs tasks headless, verifies results, and escalates on failure.

Enables AI to delegate boilerplate, drafts, tests, and refactors to free LLM providers, saving tokens and running tasks in parallel.

Enables explicit, auditable project-context handoffs between ChatGPT Work and Codex through nine MCP tools for upserting projects, creating handoffs and checkpoints, and retrieving project context and history.

Enables AI agents to access a unified catalog of tools from various APIs (OpenAPI, GraphQL, MCP, Google Discovery) through the MCP protocol.

Enables separate Claude Code sessions across a team to see teammates' live presence, warn before editing a file someone else is editing, route shared-contract changes to the developers whose code depends on them, and produce structured handoffs when a session ends.

Adds a human-in-the-loop checkpoint to MCP-capable AI coding agents, enabling them to pause and request user feedback before executing actions.

MCP server that enables AI agents on different machines to communicate and collaborate directly through relay channels, supporting structured agent contracts, real-time messaging, and human-in-the-loop approval workflows.

Diagnoses the Codex MCP launch path and writes host configuration so Codex, Cursor, or Claude Code can spawn the dsh profile. It does not run the MCP stdio server in the Web process.

Enables MCP clients to asynchronously submit tasks to a persistent coding agent already running on a development machine, disconnect, and retrieve results later through a secure, policy-controlled gateway.

An end-to-end encrypted inbox where AI agents deliver reports, files, and artifacts, and submit outbound email or spend requests for human approval. Content is decrypted in the browser, never on the server.

Enables real-time communication between Claude Code instances across multiple machines via WebSocket, allowing context sharing, task handoffs, and coordination between sessions.

A durable MCP-native job relay that enables agents to dispatch work to each other and wakes a running Claude session via a channel when a job completes, eliminating fragile tmux-based approaches.

A persistent local MCP relay that caches tool discovery, keeps upstream servers warm, and allows agents to hot-add or refresh MCP servers through management tools.

MCP client for AgentRelay, enabling local Codex agents to communicate with an AgentRelay HTTP relay for agent orchestration and task management.

MCP coordination primitive for multi-agent workflows. Lets multiple AI coding agents (Claude Code, Cursor, n8n, custom clients) discover each other and exchange messages and tasks via a shared SQLite-backed message bus.

Enables external AI applications to access MCP tools by providing a user-hosted HTTPS endpoint that forwards requests to a paired desktop client.

Enables selecting exact conversation excerpts, carrying forward only confirmed background and provenance, and handing the packaged selection to another agent for continuation.

A unified MCP gateway that reduces context token usage by exposing all services through just two tools (search and execute), supporting built-in services like Asana and Sentry as well as proxied external MCP servers.

Acts as a local stdio MCP service that lets Codex delegate implementation tasks to Antigravity CLI while conserving tokens; it returns compact review packets with changed files, diff stats, and run history for Codex to review and iterate.