API Overview

The Tensorheart Memory API is a RESTful API that uses JSON for requests and responses.

Base URL

https://api.memory.tensorheart.com/v1

Authentication

Include your API key in every request:

Authorization: Bearer mem_live_your_api_key

Endpoints Summary

Memories

Method	Endpoint	Description
POST	/memories	Create a memory
GET	/memories	List memories
GET	/memories/{id}	Get a memory
PUT	/memories/{id}	Update a memory
DELETE	/memories/{id}	Delete a memory
POST	/memories/consolidate	Merge similar memories

Memory Spaces

Method	Endpoint	Description
POST	/spaces	Create a space
GET	/spaces	List spaces
GET	/spaces/{id}	Get a space
PUT	/spaces/{id}	Update a space
DELETE	/spaces/{id}	Delete a space

Query & Retrieval

Method	Endpoint	Description
POST	/query	Find relevant memories
POST	/query/answer	Query with AI answer
POST	/query/extract	Extract memories from text

Transparent Proxy

Method	Endpoint	Description
POST	/proxy/{url}	Proxy LLM request with memory

Context Extension

Method	Endpoint	Description
POST	/chat/sessions	Create chat session
POST	/chat/sessions/{id}/messages	Send message
GET	/chat/sessions/{id}/context	Get optimized context
GET	/chat/sessions	List sessions

Document Ingestion

Method	Endpoint	Description
POST	/ingest/file	Upload and process file
POST	/ingest/url	Ingest from URL
GET	/ingest/{id}	Get document status
GET	/ingest	List documents

Tool Call Tracking

Method	Endpoint	Description
POST	/tools/calls	Record tool call
GET	/tools/calls	List tool calls
GET	/tools/summary	Usage summary

Account

Method	Endpoint	Description
GET	/account	Get account details
GET	/account/api-keys	List API keys
POST	/account/api-keys	Create API key
DELETE	/account/api-keys/{id}	Revoke API key

Usage

Method	Endpoint	Description
GET	/usage	Current usage
GET	/usage/history	Historical usage
GET	/usage/projection	Projected usage

Response Format

Success Response

{
  "success": true,
  "data": { ... },
  "meta": {
    "request_id": "req_abc123",
    "processing_time_ms": 45
  }
}

Error Response

{
  "success": false,
  "error": {
    "code": "ERROR_CODE",
    "message": "Human-readable message"
  }
}

Rate Limit Headers

Every response includes:

• X-RateLimit-Limit: Your limit
• X-RateLimit-Remaining: Remaining requests
• X-RateLimit-Reset: Reset timestamp

Pagination

List endpoints support pagination:

GET /v1/memories?page=1&per_page=50

Response includes:

{
  "pagination": {
    "total": 150,
    "page": 1,
    "per_page": 50,
    "total_pages": 3,
    "has_next": true
  }
}