jeevesagent.memory.sqlite

SQLite-backed Memory — persistent, single-file, no server.

Fills the gap between InMemoryMemory (lost on restart) and network-backed backends (Chroma / Postgres / Redis — real infra). One .db file holds every memory layer:

• working_blocks(name, content, pinned_order, updated_at) — the in-context blocks

• episodes(id, session_id, user_id, occurred_at, input, output, embedding) — episodic record + optional vector

• facts(...) via the existing SqliteFactStore rooted at the same file (separate table; same DB)

What this is good for:

• Single-instance production apps that want persistence without running Postgres / Redis.

• Local dev where you want runs to survive ctrl-c.

• CI / integration tests that need real durability without spinning up containers.

What this is NOT for:

• Concurrent writers from multiple processes — sqlite serialises writes, throughput suffers under contention. Use PostgresMemory if you have multiple workers writing to the same memory.

• Vector search at million-row scale — we do brute-force cosine ranking in Python because sqlite has no native vector type. Fine for tens of thousands of episodes; if you have more, switch to Chroma or Postgres+pgvector.

Sync sqlite3 calls are dispatched through anyio.to_thread.run_sync so the agent loop’s structured concurrency stays clean.

Classes

SqliteMemory

Durable Memory rooted at a single sqlite file.

Module Contents

class jeevesagent.memory.sqlite.SqliteMemory(path: str | pathlib.Path, *, embedder: jeevesagent.core.protocols.Embedder | None = None, with_facts: bool = True, fact_store: jeevesagent.memory.facts.FactStore | None = None)

Durable Memory rooted at a single sqlite file.

Construct directly from a path:

memory = SqliteMemory("./bot.db")
agent = Agent("...", model="gpt-4.1-mini", memory=memory)

Or via the resolver:

agent = Agent("...", model="gpt-4.1-mini", memory="sqlite:./bot.db")

Pass path=":memory:" for an ephemeral in-process database (lost on close — useful for tests).

The fact store is auto-attached: the same .db file holds a facts table managed by SqliteFactStore. Pass with_facts=False to skip it; pass an explicit fact_store= to override (e.g. point facts at a different sqlite file).

async aclose() → None

async append_block(name: str, content: str, *, user_id: str | None = None) → None

async consolidate() → None

async export(*, user_id: str | None = None) → jeevesagent.core.types.MemoryExport

async forget(*, user_id: str | None = None, session_id: str | None = None, before: datetime.datetime | None = None) → int

async profile(*, user_id: str | None = None) → jeevesagent.core.types.MemoryProfile

async recall(query: str, *, kind: str = 'episodic', limit: int = 5, time_range: tuple[datetime.datetime, datetime.datetime] | None = None, user_id: str | None = None) → list[jeevesagent.core.types.Episode]

async recall_facts(query: str, *, limit: int = 5, valid_at: datetime.datetime | None = None, user_id: str | None = None) → list[jeevesagent.core.types.Fact]

async remember(episode: jeevesagent.core.types.Episode) → str

async session_messages(session_id: str, *, user_id: str | None = None, limit: int = 20) → list[jeevesagent.core.types.Message]

async update_block(name: str, content: str, *, user_id: str | None = None) → None

async working(*, user_id: str | None = None) → list[jeevesagent.core.types.MemoryBlock]

property embedder: jeevesagent.core.protocols.Embedder

property path: pathlib.Path