Connect your AI agent to Apify
The MCP server is the live connection an agent uses to search Apify Store, run Actors, and read the results. Agent Skills carry the procedure for multi-step work, and a plugin installs both in one step for a specific coding agent. Machine-readable docs and agentic payments let an agent read the documentation and pay for runs without signing up.
To run your first Actor from an agent, follow the agent quickstart.
Pick your agent
Each card opens that agent's setup guide in Integrations.
Grok Build and Kilo Code install from the plugins page.
Other AI agents and LLM chatbots can connect directly to Apify via the MCP server.
Key concepts
With the MCP server alone, an agent can call any Apify tool but has to work out the order of steps itself. Agent Skills add the tested sequence for a task such as building an Actor. The pieces stack rather than compete, and most setups use more than one:
Build on Apify instead
The pages above cover agents that call Apify. These adjacent topics cover building on the platform:
- Develop AI agents on Apify - Build an agent that runs as an Actor on the Apify platform.
- Build Actors with AI - Use a coding agent to create and deploy an Actor.
- MCP connectors - Let an Actor you built act on a user's third-party accounts, such as Notion, Slack, or GitHub.
- AI integrations - Call Apify from agent frameworks such as LangChain, CrewAI, and the Vercel AI SDK, and load results into vector databases.
- API client libraries - Call the Apify API from your own application code, or use the JavaScript client or the Python client.
