# Deedglass

> Deedglass is the execution and accountability layer for AI agents. It sits between your agents and the tools they use: it decides what each agent may do, runs the action under control, checks the result against the system of record (L0 Reported, L1 Confirmed, L2 Witnessed by Postgres), and seals a signed deed anyone can verify offline. It is self-hosted, runs anywhere Postgres runs, and works with Claude, OpenAI, Gemini, Cursor and any MCP client.

- Start with the Quickstart, then "What VERIFIED means" for the exact meaning of a verdict.
- Every API error has its own page under /errors/<code>/.
- Every feature page ends with "What it doesn't do".

## Documentation Sets

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

## Notes

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