Api MCP Server

Connect Api to Claude, Cursor or any other MCP client and it stops being a tab you switch to. Docs · Getting started · Concepts · Slack community · Contributing. The api mcp server is what makes that connection.

What the server does

• Workflows as code. — YAML under .shipfox/workflows/, versioned and reviewed
• Trigger from your whole stack. — Start runs from GitHub, Sentry, Slack, Linear,
• Automatic retry loops. — Guard a step with a check (a gate): when it fails,
• Long-running, event-driven agents. — A listening job stays alive across a run
• See the agent think. — Agent steps stream structured events (messages,
• Pick the agent, not the model. — Run an agent step on the pi harness

Available tools

The toolset is worth reading before you wire it up, because it tells you what the integration is really for:

• run — npm install
• key — fix # an agent step
• Long — running, event-driven agents.** A listening job stays alive across a run
• Slack — ** join the Shipfox community Slack
• Issues — ** report bugs and request features on
• Concept — Summary
• Workflow — A YAML file under .shipfox/workflows/, versioned and reviewed like code. One file, one workflow
• Trigger — What starts a run: an event from a connected integration, or an on-demand fire
• Integration — A connection to an external tool (GitHub, Sentry, Slack, Linear, ...) whose events start runs. Use the [generic
• Job — A group of steps on one runner. Jobs form a DAG via needs and are isolated, so each re-clones the repo
• Step — A run shell command or an agent (model + prompt, on the pi or claude harness). Runs in order within a job
• Gate — A pass/fail check on a step that retries from an earlier step when it fails. Bounded retry

Installation

Setup follows the usual MCP pattern — install or clone the server, register it in your client's configuration file, restart the client.

Where it fits

This sits in the team communication group, where several servers overlap in what they claim to do but differ sharply once you actually set them up. Api's toolset — run, key, Long and 10 more — is a fair guide to whether it matches your workflow. It is maintained by GitHub Actions; worth a glance at recent repository activity before you build anything load-bearing on it.

This entry was verified against Api's own documentation before publication; SyncDev keeps the directory reviewed rather than auto-generated.

Worth knowing first

• It runs with your machine's permissions. That is convenient and also the reason to think about what you point it at before you approve a tool call.
• With 13 tools registered it takes up a noticeable share of the context window; turn it off in projects that never touch Api.
• Keep per-call confirmation enabled while you learn its behaviour; it is the cheapest safeguard you have.