Integrations

Your library, wherever the work happens.

Your own AI reads the library directly — every photo with its date, its place and a written description of what is in it. No export, no folder to watch, no connector to wait for.

The API

An API your agent can query.

If you already run an agent, it does not need us to integrate with it. It calls us. A bearer token gets you the three read calls; writing results back takes a second scope on the same token. Anything that can make an HTTP request can do all four.

Every call is authenticated, scoped to one workspace, and recorded in an audit table. No call ever crosses from one workspace into another.

Works with

Claude, ChatGPT, and whatever you use next.

There is nothing to install and no per-vendor connector to wait for. Anthropic’s Claude, OpenAI’s models, an OpenRouter setup, a script you wrote yourself — if it can call an HTTP endpoint with a bearer token, it can read your library today. That is deliberate: a connector per vendor is a connector to maintain per vendor, and you would be waiting on us every time you changed model.

Storage

Keep it here, or keep it everywhere.

The library lives with us, which is where the descriptions and the search live too. If your company already pays for storage, mirror it there as well and nothing changes about how you ask for a photo.

Access

Point it at your own events.

Tell us what your events look like and what you would want an agent to do with the photos, and we will tell you whether this fits and what it costs.

Request access