Knowledge is the surface that exposes what cf0 knows — every topic it has compiled, every note that anchors a claim, every source it has ingested, and the graph that connects them. It’s the audit and inspection layer behind every Lab answer. Open Knowledge from the left sidebar.Documentation Index
Fetch the complete documentation index at: https://docs.cf0.ai/llms.txt
Use this file to discover all available pages before exploring further.
The ten sub-views
Topics
The compiled view — every entity cf0 tracks (companies, people, deals, sectors) grouped by type. The default landing view of Knowledge.
Entity detail
Drill into a single entity (e.g. a specific ticker or person) to see every note, source, and connection attached to it.
Notes
Structured facts cf0 has extracted — one note per claim, with the source it came from, the date, and the topic it belongs to.
Sources
Every channel feeding the knowledge surface — SEC filings, transcripts, Slack channels, Gmail labels, Notion pages, uploads. Filter by channel.
Graph
Force-directed map of topics and the edges between them. Hover a node for the supporting notes. Useful for finding non-obvious connections.
Inbox
New material cf0 has ingested but not yet routed to a topic. Review and approve or recategorise.
Routines
The scheduled reads that keep Knowledge fresh — ingestion jobs, summarisation passes, drift detection. See last-run status and next-run time per routine.
Audit
Append-only log of every change to a note or topic — who added it, what changed, which source it came from. Compliance-grade.
Dreams
Speculative connections cf0 has surfaced but isn’t yet confident enough to promote into the compiled view. Review and accept or dismiss.
Admin
Org-admin controls for Knowledge — managing sources, pruning topics, configuring routines. Admins only.
How Knowledge feeds Lab
Every Lab thread reads Knowledge before it answers. When you ask about a name, Lab pulls relevant topics, the notes that anchor them, and the sources behind each note — alongside the SEC filings and live market data. That’s why a question about a portfolio company can pull from a Granola meeting transcript and a 10-Q in the same answer: both flow through the Knowledge surface.Provenance, end-to-end
Source arrives
cf0 reads from a connected integration, an uploaded document, or a scheduled SEC ingest. The source is registered under Sources with a channel, a timestamp, and the ingest job that fetched it.
Notes extracted
A routine extracts structured notes from the source. Each note links back to the exact source passage it came from.
Notes routed to topics
Notes are tagged with the entities they reference (companies, people, deals). The Topics view aggregates notes by entity.
Lab reads from topics
When you ask about an entity in Lab, the model receives the topic’s notes — and their sources — as context. Every answer can resolve back to its source passage.
Personal vs org scope
Knowledge has two scopes:| Scope | Sources | Notes | Visible to |
|---|---|---|---|
| Personal | Your uploads, your connected integrations | Notes you’ve authored or annotated | You only |
| Org | Approved org documents, org-connected integrations | Notes promoted to org scope | Every member of the org |
Org scope is editable only by org admins or members who have been granted Knowledge write access. Members can always read the org scope.