mule-apilisted

# Mule AI Agent Skill This skill helps you interact with a running Mule AI workflow platform. Mule is an AI workflow platform that enables you to create, configure, and execute AI agents and workflows through an OpenAI-compatible API. ## Prerequisites ### Server Connection Before interacting with Mule, you must determine the server URL: 1. **Check for MULE_SERVER environment variable**: ```bash echo $MULE_SERVER ``` 2. **If MULE_SERVER is not set**, ask the user to either: - Set the `MULE_SERVER` environment variable (e.g., `export MULE_SERVER=https://mule.butler.ooo`) - Provide the Mule server URL directly 3. **All API requests** should be made to the base URL stored in `MULE_SERVER` (e.g., `curl ${MULE_SERVER}/v1/models`) ### Base URL Variable Throughout this skill, replace `${MULE_SERVER}` with the actual server URL. Common examples: - `http://localhost:8080` (local development) - `https://mule.butler.ooo` (production) --- ## Overview of Mule Primitives Mule uses six core primitives stored in PostgreSQL: 1. **Providers** - AI provider configurations (OpenAI-compatible APIs) 2. **Skills** - Pi agent skills that can be assigned to agents 3. **Agents** - AI agents powered by pi RPC runtime 4. **WASM Modules** - WebAssembly modules for imperative code 5. **Workflows** - Ordered sequences of workflow steps 6. **Workflow Steps** - Individual execution steps (AGENT or WASM type) --- ## Listing Available Models To see what agents and workflows are av