Mule API

What it does

Mule API skill lets an agent interact with a running Mule AI workflow platform to list models (agents and workflows), execute agents or workflows (sync or async), manage providers and skills, and monitor jobs. It provides concrete curl examples and step-by-step patterns for creating agents, workflows, and WASM modules, and for running and tracking jobs.

When to use it

Use this skill when you need to orchestrate complex AI workflows hosted on a Mule server: creating or updating providers, spinning up agents, invoking workflows (both synchronous and asynchronous), or querying job status. It’s suited for environments where Mule exposes an OpenAI-compatible API endpoint (MULE_SERVER).

What's included

• Scripts: none bundled, but detailed curl examples for all API flows are provided
• References: inline API endpoint examples and job/workflow management patterns
• Instructions: how to discover models, execute agents/workflows, manage providers, create agents and skills, and work with async job lifecycle. The skill documents common error codes and WebSocket usage for real-time job updates.

Compatible agents

Compatible with agents and runtimes that can call HTTP endpoints and execute shell curl commands (e.g., Copilot/Codex, CLI-based agents, or any agent with HTTP tooling).