Announcing Polygraph

Every once in a while you see a real problem and land on a good solution. Ten years ago, that was monorepos. Last year, we found another one. We felt it ourselves, and we heard it from clients building ambitious applications with AI.

The pain: agents can't work autonomously inside an organization. An agent can write a Rust compiler from scratch, but it can't update a button label and ship the change without several people unlocking the process.

I talked to dozens of teams at the AI Engineer New York conf. Every one recognized the problem. Some were already building internal tooling to fix it.

We knew we could fix it, so we built Polygraph.

Polygraph is an agent-agnostic meta-harness. If "meta-harness" doesn't mean anything to you, think of a meta-framework like Next.js. Next.js adds the missing pieces that make React useful for building apps. Polygraph adds the missing pieces that make agents autonomous in real workflows.

Agents hit two walls:

• Space. An agent is stuck in one repo. It can't see how a change fits the wider system, and it can only write to one repo at a time.
• Time. An agent doesn't have any episodic memory. Every session starts blank, so a human carries the memory context.

Both walls cap autonomy. Developers coordinate changes across every affected repo, run multiple sessions to land one simple change, and re-explain the same context on every PR.

Polygraph removes both:

• It connects all your repos, private and public, into one dependency graph without moving any code. Your agent reads and writes across every repo you can access and orchestrates the PRs and CI as a single change.
• It records every session, so you can resume, reference, or build on work done by anyone, on any machine, even with a different agent.

With Polygraph, your agents see every repo in your organization and know who is doing what work, and how. Every change, every decision, every session becomes part of your Polygraph.

Try It Today

Register at trypolygraph.com

Learn More