Account.

Developer plan.

API keys.

Keys are shown once, when you create them. Store them in a secret manager, never in client-side code.

Conversations are used to teach Jesse; what you tell her about yourself stays with your account.

Jesse is a Bayesian predictive model of validated priors on a shared root, jesse-pristine. The skills and general facts she learns from conversations go to that shared model, each with its provenance and label. Your key adds a layer of your own on top of it: the facts you state, the corrections you send, what your searches teach, and the tool packs you store. Your library and Agents.md belong to the account, for every key. Your private content is never shown in another user's answers. GET /v1/agent shows what a key has learned; POST /v1/agent/reset puts it back, and data you ask her to forget is never learned from again.

Create key

curl https://jesse.my/api/v1/chat/completions \
  -H "Authorization: Bearer $JESSE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "jesse", "messages": [{"role": "user", "content": "What is 17 * 23?"}]}'

Your library.

Files you upload in chat are saved here, so Jesse can use them later and doesn't read the same file twice. Messages aren't saved. Delete a file to remove it everywhere.

No files yet.

Meanings Jesse remembers.

When a word has several meanings and you pick one, Jesse counts it, so next time it answers with your usual meaning and checks with you when a conversation points the other way. Only the word and the meaning are kept.

None yet.