Attestor
@bolnet
Audit-grade memory backbone for agent teams. Bi-temporal facts (event time + transaction time, with recall(as_of=...) replay), 6-step deterministic retrieval (no LLM in the critical path), conversation ingest with speaker-locked dual-pass extraction, per-tenant Postgres row-level
Overview
What is Attestor?
Attestor is a memory store for agent teams that need a shared, tenant-isolated memory with bi-temporal replay, deterministic retrieval, and an auditable supersession chain. It runs as a Python library, a Starlette REST service, or an MCP server β the same API in all three.
How to use Attestor?
Install via pip install attestor, set up local Postgres and Neo4j using attestor setup local, pull the default embedder (ollama pull bge-m3), then verify with attestor doctor. Use the Python API (AgentMemory, AgentContext) or run as an MCP server.
Key features of Attestor
- Bi-temporal memories with event and transaction time axes for point-in-time reconstruction.
- Deterministic six-step retrieval pipeline with no LLM in the hot path.
- Tenant isolation via Postgres Row-Level Security.
- Conversation ingest with two-pass speaker-locked
