apify-automation

Apify Automation

Run Apify web scraping Actors and manage datasets directly from Claude Code. Execute crawlers synchronously or asynchronously, retrieve structured data, create reusable tasks, and inspect run logs without leaving your terminal.

Toolkit docs: composio.dev/toolkits/apify

---

Setup

1. Add the Composio MCP server to your configuration:

   https://rube.app/mcp
   

2. Connect your Apify account when prompted. The agent will provide an authentication link.
3. Browse available Actors at apify.com/store. Each Actor has its own unique input schema -- always check the Actor's documentation before running.

---

Core Workflows

1. Run an Actor Synchronously and Get Results

Execute an Actor and immediately retrieve its dataset items in a single call. Best for quick scraping jobs.

Tool: APIFY_RUN_ACTOR_SYNC_GET_DATASET_ITEMS

Key parameters:

• actorId (required) -- Actor ID in format username/actor-name (e.g., compass/crawler-google-places)
• input -- JSON input object matching the Actor's schema. Each Actor has unique field names -- check apify.com/store for the exact schema.
• limit -- max items to return
• offset -- skip items for pagination
• format -- json (default), csv, jsonl, html, xlsx, xml
• timeout -- run timeout in seconds
• waitForFinish -- max wait time (0-300 seconds)
• fields -- comma-separated list of fields to include
• omit -- comma-separated list of fields to exclude

Example prompt: "Run the Google Places scraper for 'restaurants in New York' and return the first 50 results"

---

2. Run an Actor Asynchronously

Trigger an Actor run without waiting for completion. Use for long-running scraping jobs.

Tool: APIFY_RUN_ACTOR

Key parameters:

• actorId (required) -- Actor slug or ID
• body -- JSON input object for the Actor
• memory -- memory limit in MB (must be power of 2, minimum 128)
• timeout -- run timeout in seconds
• maxItems -- cap on returned items
• build -- specific build tag (e.g., latest, beta)