API Documentation

📖 Overview

The AI Memory System API allows you to programmatically manage projects, memories, sessions, tasks, and code snippets. All endpoints return JSON responses and accept JSON request bodies where applicable.

🔑 Authentication

All API requests require authentication using a token. Include your token in the Authorization header of each request.

curl -X POST http://aimemory.com.br/api/auth/login/ \
  -H "Content-Type: application/json" \
  -d '{
    "username": "your_username",
    "password": "your_password"
  }'

{
  "token": "abc123def456...",
  "user_id": 1,
  "username": "your_username",
  "email": "user@example.com",
  "tenant": {
    "id": "uuid-here",
    "name": "Your Tenant",
    "slug": "your-tenant",
    "plan": "premium"
  }
}

🚀 Quick Start

curl -X POST http://aimemory.com.br/api/memories/ \
  -H "Authorization: Token YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "project": "your-project-slug",
    "memory_type": "decision",
    "content": "We chose PostgreSQL for better JSON support",
    "importance": 8,
    "tags": ["database", "architecture"]
  }'

📁 Projects

Projects are containers for organizing memories, sessions, tasks, and code snippets. Each project has a unique slug for easy reference.

[
  {
    "id": "uuid",
    "name": "My Project",
    "slug": "my-project",
    "description": "Project description",
    "stack": "Django, React, PostgreSQL",
    "is_active": true,
    "created_at": "2024-01-01T00:00:00Z",
    "updated_at": "2024-01-01T00:00:00Z"
  }
]

curl -X POST http://aimemory.com.br/api/projects/ \
  -H "Authorization: Token YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "name": "My New Project",
    "description": "A cool project",
    "stack": "Python, Django, React"
  }'

🧠 Memories

Memories are the core data type. Store facts, decisions, learnings, errors, solutions, and code notes. Each memory has an importance level (1-10) and can be tagged for organization.

[
  {
    "id": "uuid",
    "project": "uuid",
    "memory_type": "decision",
    "content": "We chose PostgreSQL for better JSON support",
    "context": "Database selection meeting",
    "importance": 8,
    "tags": ["database", "architecture"],
    "access_count": 5,
    "last_accessed": "2024-01-15T10:30:00Z",
    "created_at": "2024-01-01T00:00:00Z"
  }
]

curl -X POST http://aimemory.com.br/api/memories/ \
  -H "Authorization: Token YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "project": "my-project",
    "memory_type": "decision",
    "content": "We chose PostgreSQL for better JSON support",
    "context": "Database selection meeting",
    "importance": 8,
    "tags": ["database", "architecture"]
  }'

📅 Sessions

Sessions track your work periods with AI. Record what you accomplished, blockers encountered, and next steps.

curl -X POST http://aimemory.com.br/api/sessions/ \
  -H "Authorization: Token YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "project": "project-uuid",
    "summary": "Implemented user authentication",
    "accomplished": [
      "Created login endpoint",
      "Added JWT token validation"
    ],
    "next_steps": [
      "Add refresh token mechanism",
      "Implement password reset"
    ],
    "blockers": [
      "Need Redis for token storage"
    ]
  }'

✅ Tasks

Track tasks and to-dos for your projects. Set priorities, due dates, and manage status.

curl -X PATCH http://aimemory.com.br/api/tasks/TASK_UUID/ \
  -H "Authorization: Token YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"status": "completed"}'

💻 Code Snippets

Store important code fragments with metadata like file path, language, and line numbers.

📋 Context Generation

Generate comprehensive context documents for your projects. Perfect for starting new AI conversations with full project knowledge.

curl http://aimemory.com.br/api/projects/PROJECT_UUID/context/ \
  -H "Authorization: Token YOUR_TOKEN"

⚠️ Error Handling

The API uses standard HTTP status codes and returns JSON error responses.

{
  "detail": "Authentication credentials were not provided."
}

📄 Pagination

List endpoints support pagination using limit and offset parameters.

GET /api/memories/?limit=20&offset=40

🔌 MCP Integration

AI Memory System integrates with Claude Code via the Model Context Protocol (MCP). This allows Claude to directly access and manage your memories during coding sessions.

{
  "mcpServers": {
    "ai-memory": {
      "command": "python",
      "args": ["/path/to/mcp_server/server.py"],
      "env": {
        "AIM_API_URL": "http://aimemory.com.br"
      }
    }
  }
}