Two stdio MCP servers — weather forecast and invoice approval — each backed by a registered Resonate generator workflow.
Wrap a Resonate workflow behind a stdio MCP server so an AI client can call it as a tool. Each tool invocation runs as a durable execution with replay, retry on throw, and durable timers.
Three MCP tools coordinate on one durable promise keyed by job name — work survives MCP-server restarts and deduplicates by id.
Three MCP tools — start_gathering, probe_status, await_result — coordinate on one Resonate durable promise keyed by job name; after a crash the worker reclaims the promise and re-runs the function from the top.
How an MCP tool call dispatches into a Resonate workflow that checkpoints each step and deduplicates by promise ID.
Expose a durable Resonate workflow as a Model Context Protocol tool over stdio. The MCP server stays stateless, the workflow checkpoints each step on a worker process, and concurrent calls for the same input deduplicate by promise ID.
How an MCP server returns a promise id instead of blocking, and lets the agent ask back later — backed by a Resonate-registered function and a durable sleep.
Expose a long-running job to an LLM as an MCP tool that returns a promise id immediately, then let the agent poll for completion against a durable promise on the Resonate server.
Polling loops, deploys, DNS checks, image-generation jobs — when durable shell is the right primitive.
Resonate's `resonate-bash` MCP tool runs shell commands as durable, asynchronous tasks. Good at: long polling loops, operations that outlive the session, named queryable state, and fire-and-watch coordination with external systems.