BriefFactoryOS

Reference Demo

BriefFactoryOS

AI Competitive Intelligence Platform

An AI competitive-intelligence pipeline that reads 41 public sources on any company and writes a sourced brief with ROI. The AI model assists and audits, and every figure traces to a public record.

Demo walkthrough · 1:16 Companies shown are for demonstration purposes only. Read the disclaimer.

BriefFactoryOS turns public records into a sales conversation. Point it at a company and it reads 41 free public sources, matches what it finds against a catalog of 58 AI use cases, and writes an internal brief for pitching which of those use cases actually fit, with ROI ranges where public data or a benchmark supports one.

The AI model assists and audits. Findings, matching, and ROI stay deterministic and auditable, while the model drafts revisions and runs an automated judge that re-checks every claim against its stored source. Every figure traces to a source URL and a retrieval date, and the brief stays a draft until a human reviews it.

Built with Claude Code.

41
Free public data sources
58
AI use cases in the catalog
47
Automated test suites
~52K
Lines across 20 schema migrations
Under the hood

How it is built

01

Public sources in, client-ready brief out

Profiles any company from 41 free public data sources (SEC EDGAR, federal courts, the Federal Register, FDA, USPTO, state labor and privacy registries), matches findings against a 58-entry AI use-case catalog, and generates a client-ready brief with ROI ranges.

02

The AI model assists and audits

Findings, matching, and ROI stay deterministic and auditable, while the model drafts revisions and runs an automated judge that re-checks every claim against its stored source.

03

A traceability contract

Every claim carries a source URL and retrieval date, and open questions surface as “unknown, flagged for discovery.” Backed by 47 automated test suites, including guards that fail when a number and its citation disagree.

04

The full stack

Python/FastAPI, React/TypeScript, Postgres on Supabase, ~52K lines across 20 schema migrations. Developed with Claude Code as an AI pair programmer.

PythonFastAPIReactTypeScriptPostgresSupabaseClaude Code
Disclaimer

About this demo

Any companies, names, logos, or trademarks that appear in the demo are used for demonstration purposes only. Their appearance does not imply any affiliation, endorsement, sponsorship, or client relationship with those companies, and they have not reviewed or approved this content.

Information in the demo is drawn from free public sources as of the retrieval dates shown and may be incomplete, out of date, or inaccurate. Findings, use-case matches, and ROI ranges are illustrative outputs of an automated pipeline. They are not statements of fact about, or assessments of, any company.

Nothing on this page or in the demo is legal, financial, or investment advice. All trademarks are the property of their respective owners.