Temporal MCP Server

This app is a Model Context Protocol (MCP) server that acts as a bridge between AI assistants (like GitHub Copilot, Claude, or OpenCode) and a Temporal workflow orchestration cluster. Rather than writing code or using the Temporal CLI manually, you can interact with your workflows entirely through natural language, asking your AI assistant to start a workflow, check its status, send it a signal, cancel it, or inspect its full event history. The server exposes 19 tools covering the full Temporal surface area: individual workflow lifecycle management, real-time querying and signaling, batch operations across many workflows simultaneously, and complete schedule management via cron expressions.

The problems it solves are primarily around operational friction and observability. In production environments, Temporal workflows often need ad-hoc intervention such as retrying a stuck workflow, bulk-canceling a bad batch, inspecting why a particular execution failed, or triggering a scheduled job early for testing. Normally this requires CLI access, familiarity with Temporal's query syntax, or navigating the Temporal UI. This server makes those operations conversational and accessible to anyone with an AI coding assistant, regardless of their familiarity with Temporal internals. It supports local deployments as well as Temporal Cloud, with both mTLS client certificate auth and API key auth built in, making it suitable for production cloud environments out of the box.