# Golemry Docs

> Tell your agent what you want done. Golemry turns it into an automation that runs on a schedule or trigger in the cloud, with an overseer reviewing every result.

Key facts for reading these docs:

- You connect Golemry to an AI assistant you already use (your **main agent**) as a remote MCP server at `https://mcp.golemry.com/mcp`, and manage automations by chatting with it. Everything can also be done directly in the Golemry web app.
- An automation is a **job**: a prompt, a model, and optional connectors and skills.
- A job runs when its schedule fires, when a connector event triggers it, or when it is run manually.
- Each run executes in an isolated cloud sandbox, carried out by the **worker agent**.
- After each run an **overseer** agent reviews the result and records a verdict of passed, flagged, or failed.
- Users bring their own LLM API key. Model usage is billed by that provider, not by Golemry.

## Documentation Sets

- [Abridged documentation](https://docs.golemry.com/llms-small.txt): a compact version of the documentation for Golemry Docs, with non-essential content removed
- [Complete documentation](https://docs.golemry.com/llms-full.txt): the full documentation for Golemry Docs

## Notes

- The complete documentation includes all content from the official documentation
- The content is automatically generated from the same source as the official documentation

## Optional

- [Integrations directory](https://golemry.com/integrations): Every available connector, with the tools and event triggers each one offers.
- [Pricing](https://golemry.com/pricing): Plans and the monthly task allowance each one includes.
- [Data and security](https://golemry.com/security): How credentials, run data, and subprocessors are handled.