Prototype tool
Referent
AI agents now write code faster than humans can review it line by line. The unit of review has to move up — from diffs to capabilities, from files to boundaries.
A context engine for software systems. Versioned, deterministic, bounded — so humans and AI agents can change code from the same grounded understanding.
AI agents now generate code at a scale humans cannot review line by line. Approving ten-thousand-line pull requests by gut feel is how production incidents, security holes, and compliance failures get shipped. The reviewer has the responsibility but no longer has the surface area.
Referent is an attempt at the surface area. It reads a codebase deterministically and produces a versioned graph of what is actually there — capabilities, paths, boundaries, ownership — so the unit of review can move from lines to what changed in the system.
What it explores
- Truth before explanation. A static, deterministic extraction layer first. Anything an LLM contributes is grounded in code that was already identified by the parsers — never invented.
- One substrate, two surfaces. A graph for humans to navigate, and bounded context packs for agents to consume. Both are projections of the same versioned snapshot, so they cannot drift apart.
- Bounded by construction. Token budgets are hard ceilings, not aspirations. A context pack is reproducible, citable, and diff-able against a prior pack.
- Capability-level diffs. The point of the graph is the diff. Did checkout suddenly start calling billing directly? Did the auth boundary move in this PR? Which paths touch sensitive data this week that didn’t last week?
Why it matters
Most AI coding failures are context failures. Agents are given partial, lossy, stale context and then asked to make changes with incomplete understanding. Most reviewer failures are the inverse: too much context, presented at the wrong altitude. Both problems share a substrate.
The bet is that the codebase itself — parsed, versioned, made navigable — is a more honest source of truth about a running system than any document, diagram, or runtime dashboard, and that a small, disciplined layer over it is enough to change how AI-written software is reviewed and shipped.
Status
In active development. Local-first by design. Public details will follow as the work matures; the underlying ideas show up in the writing.