Knowledge & Insights
Discover what Fabriq learns from your codebase — structured knowledge, feature history, and audit trails.
Beyond managing features, Fabriq continuously learns from your codebase. It extracts structured knowledge — behaviors, business rules, state machines, evidence — and makes them navigable for both humans and agents.
Your codebase is more than source files. It's a map of decisions, patterns, and conventions. Fabriq's Knowledge section helps you explore that map.
Knowledge types
| Type | What it contains | How it's created |
|---|---|---|
| CDR Assets | Behaviors, business rules, state machines, evidence extracted from code | CDR pipeline scan |
| Feature History | Decision records, close summaries, promoted docs from closed features | Feature close pipeline |
| Audit Trail | Every capability call, confirmation, and stage advancement | Automatic, append-only log |
Why it matters
- For engineers: Browse the cognitive index instead of re-scanning the codebase manually. See what business rules exist before modifying code.
- For tech leads: Audit trail provides "who approved what and when" for every agent action. Feature history preserves decisions for future reference.
- For agents: The cognitive index is injectable context — agents can query structured knowledge instead of guessing.
Next steps
- CDR Overview — how Fabriq extracts structured knowledge
- Running Your First CDR Scan — step-by-step tutorial
- Audit Trail — every decision, recorded
Edit on GitHub
Last updated on