One thing I’m now adding to personal project is an agent manifest.

The idea is simple: when an agent lands on the product, it should understand what the product does, what state the account is in, what actions are available, what the next best step is, and what guardrails apply.

Not just for fully autonomous agents either. Also for humans working with agent support.

I think more products will need to become legible to agents in the same way they became mobile-friendly and API-friendly.

That means moving beyond “here’s the UI” or “here’s the API” into something more explicit:
here’s the workflow, here’s the state, here’s what’s safe to do next.

Feels very aligned with the broader dark factory way of building:
clear contracts, explicit state, safe actions, fewer guesses.

Attached is a simple example of what the agent-readable output might look like.