Skip to main content

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.

cf0 is the financial intelligence stack. Three layers, one moat. Inputs read everything your firm reads: SEC filings, earnings transcripts, customer comms, internal docs. Brain compounds those inputs into a wiki that knows your customers, deals, and decisions — and remembers across sessions. Lab is where you ask it anything, with agents pulling from the brain and on-demand UI rendering the answer in seconds. Each layer is a defensible product on its own. The compounding effect — Inputs feeding Brain feeding Lab that ships cited reports — is the part competitors can’t replicate by gluing tools together.

​
The three layers

01 · Inputs

SEC EDGAR (10-Ks, 10-Qs, 8-Ks, S-1s, proxies), earnings transcripts, customer comms via Slack/Gmail/HubSpot, internal docs in Notion/Drive, news and research feeds. Everything your firm reads or writes about a name flows in, structured.

02 · Brain

A world-model that compounds. Tracks companies, people, deals, decisions and their rationale, citations, warnings, suggestions, activity, and provenance. Each new source sharpens the next answer.

03 · Lab

Ask it anything. Agents pull from the brain, generate the on-demand UI matching the question, and ship cited reports your principals can read in 30 seconds. No dashboards. No templates.

​
What ships from Lab

Lab

AI chat for financial research. Ask in plain English, get back structured tables, charts, and citations — all in one streaming thread.

Reports

Branded equity research PDFs with charts, financial models, and inline citations. Generated from a Lab thread; editable in place; exportable for distribution.

Filings

Deep SEC coverage with structured 10-K / 10-Q / 8-K extraction, plus directory data across the major global markets your team reaches into. Jump to the exact section a figure came from.

Documents

Drop in PDFs, spreadsheets, decks, images. cf0 parses and indexes them so Lab can answer questions across your private docs alongside public data.

Skills

Reusable prompt workflows for your firm’s recurring research moves — DCF, comps, earnings recaps, IC memos. Invoke with /skill-name in Lab.

​
How a research session runs

  1. Ask in Lab. Type a question in plain English — for example, “Compare NVDA and AMD gross margin trajectory over the last eight quarters.” Lab routes the query through the brain, pulls the relevant filings and transcripts, and streams structured output: chart, table, narrative.
  2. Verify every number. Each figure links back to its source — the exact filing, page, and section. Click to inspect provenance directly.
  3. Generate a report. When the thread is tight enough to share, ask Lab to generate a report. cf0 assembles a branded PDF with sections, charts, and inline citations.
  4. Encode it as a Skill. If the workflow is one you’ll run again, save it as a Skill. Your whole team can run it with one slash-command.
cf0 never trains on your data and never shares it across organisations. Filings, documents, and reports are scoped to your org workspace. Access controls run at the org level via Clerk.