Documents
The documents that matter, readable by your AI.
Each `document` is a titled record with a type — policy, spec, brief — and its full text as the body. Everything is searchable and linked from the moment it is saved.
Extraction is the second step, same as Notes: run it over a document and its facts become reviewed structured memory, connected to the people and projects they mention.
What it does
- Typed documents: policies, specs, briefs, references.
- Full text kept on the record, its own page and URL.
- One-click extraction into reviewed structured facts.
- Agents read the same documents over MCP.
What it stores
As ordinary objects in your workspace memory — readable by your AI over MCP, browsable like any memory.
- document — A kept document — doc_type says what kind, the full text is the description.
Sign in to install Documents into your workspace. Sign up — personal use is free.