BoxBrain

BoxBrain is an IdentityDB-backed TypeScript framework for creating synthetic personas that can behave like human-like DM contacts.

The project is framework-first rather than product-first. The current foundation provides:

• provider-agnostic text, structured-output, and image adapter contracts
• one IdentityDB memory space per persona
• persona initialization from personality, history, values, preferences, and relationships
• LLM-generated biography ingestion into IdentityDB fact drafts
• optional profile image generation through an image adapter
• human-like typing and first-reply delay utilities

Planned next APIs include:

• schedule generation and availability state persistence
• inbound DM-style conversation turns with mandatory/contextual memory retrieval
• proactive outbound messages without user input
• HTTP/RPC wrappers around the core library APIs

Development

bun install
bun run test
bun run check
bun run build

Current status

The repository is in foundation development. See the implementation plan:

• docs/plans/2026-05-11-boxbrain-foundation.md