Changelog

All notable changes to the Tensorheart Memory API.

v1.1.0 (2024-12-21)

Added

Smart Updates

• Intelligent conflict resolution when adding memories
• Automatic detection of contradictions, refinements, additions, and duplicates
• smart parameter on POST /v1/memories
• Resolution details included in response

Temporal Versioning

• Memory history tracking with version chains
• Point-in-time queries with as_of parameter
• GET /v1/memories/{id}/history endpoint
• Fields: valid_from, superseded_at, supersedes_id

Entity Extraction

• Automatic extraction of people, organizations, locations, skills, and products
• extract_entities parameter on POST /v1/memories
• New Entities API for querying extracted entities:
  • GET /v1/entities - List all entities
  • GET /v1/entities/{type} - List by type
  • GET /v1/entities/{type}/{name}/memories - Get memories for entity

API Endpoints

• GET /v1/memories/{id}/history - Get memory version history
• GET /v1/entities - List entities
• GET /v1/entities/{type} - List entities by type
• GET /v1/entities/{type}/{name}/memories - Get memories containing entity

---

v1.0.0 (2024-01-15)

Added

• Initial release of the Tensorheart Memory API
• Memory CRUD operations
• Query and retrieval
• Answer generation
• Memory extraction from text
• Multi-tenant organization support
• API key management with scopes
• Usage tracking and billing
• Rate limiting by plan tier

API Endpoints

• POST /v1/memories - Create memory
• GET /v1/memories - List memories
• GET /v1/memories/{id} - Get memory
• PUT /v1/memories/{id} - Update memory
• DELETE /v1/memories/{id} - Delete memory
• POST /v1/query - Query memories
• POST /v1/query/answer - Query with AI answer
• POST /v1/query/extract - Extract memories
• GET /v1/account - Account details
• GET /v1/usage - Usage metrics

---

Upcoming

Planned Features

• Webhooks for memory events
• Batch API for async operations
• Advanced analytics dashboard
• Knowledge graph relationships

---