Beads gives AI assistants persistent memory with an SQLite plus JSONL store, so teams cut context waste and deliver sooner.

The tool will use an LLM (gpt-4o-mini by default) to generate the appropriate SQL query by passing through your question and the database schema, and will then execute the query and return the result.

CREATE TABLE [conversations] ( [chat_id] TEXT PRIMARY KEY, [title] TEXT, [create_time] TEXT, [update_time] TEXT ); CREATE TABLE "messages" ( [message_id] TEXT PRIMARY ...