Explicit memory writes
Agents call remember with a raw memory. No scraping, no transcript ingestion — only what you choose to store.
Automatic classification
A cheap decision model proposes scope, sensitivity, domain and tags, with confidence scores and a needs-review queue.
Deterministic authorization
One pure function decides reads. AI output is metadata only and can never grant access.
Versioned, correctable memory
Corrections create immutable new versions. Scope and sensitivity changes are auditable policy actions.
Hybrid, scoped recall
Vector, keyword and metadata candidates are authorized before reranking. You only ever see what you may read.
OAuth connections
MCP clients connect with OAuth 2.1 and act as the authenticated human — never as a privileged robot.
Built for agents, governed for people
A client remembers something. MemScope classifies it, applies your policy ceiling, and stores it. Later, any client recalls — and receives only the memories its human principal is permitted to see.
- Raw memory is immutable per version
- Forget is a soft delete; restore is authorized
- Tenant isolation on every query
- Memory content never appears in logs
Interfaces & tools
- MCP endpoint
- mcp.memscope.ai/mcp
- REST API
- api.memscope.ai
- Tools
- remember · recall · explain · correct · forget