{
  "id": "page:docs-supermemory-research-raw-01-docs-index",
  "_kind": "Page",
  "_file": "wiki/docs/supermemory-research/raw/01-docs-index.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/supermemory-research/raw/01-docs-index.md",
    "sourceKind": "repo-docs",
    "title": "Supermemory Documentation Index",
    "displayName": "Supermemory Documentation Index",
    "slug": "docs/supermemory-research/raw/01-docs-index",
    "articlePath": "wiki/docs/supermemory-research/raw/01-docs-index.md",
    "article": "\n# Supermemory Documentation Index\n\nSource: https://supermemory.ai/docs/\n\n## Core Concept\n\nSupermemory functions as infrastructure for AI agent memory, enabling \"perfect recall about users\" to create more intelligent and personalized systems. The platform achieves state-of-the-art performance on benchmarks including LongMemEval and LoCoMo.\n\n## Key Components\n\n- Agent memory systems\n- Content extraction capabilities\n- Connectors and data syncing\n- Managed RAG platform\n\n## How It Operates\n\nThe system accepts diverse input formats -- text, files (PDF, images, documents), conversations, and video content. Supermemory then \"intelligently indexes them and builds a semantic understanding graph\" around entities like users or projects, retrieving contextually relevant information during queries.\n\n## Three Context Addition Methods\n\n1. **Memory API** -- Extracts and evolves user facts in real-time, handling knowledge updates and temporal shifts to generate user profiles\n\n2. **User Profiles** -- Combines static information (always-known facts) with dynamic, episodic details from recent conversations\n\n3. **RAG Search** -- Advanced semantic retrieval featuring metadata filtering and contextual chunking\n\nAll three approaches utilize the same context pool when sharing a user ID, allowing flexible implementation strategies.\n\n## Documentation Sections\n\n- Getting Started / Quickstart\n- Authentication (API keys, scoped keys, connector branding)\n- Introduction\n- Content Management (documents, memories, search)\n- Graph Memory (automatic memory evolution, knowledge updates, intelligent forgetting)\n- User Profiles\n- Connectors (GitHub, Gmail, Google Drive, Notion, OneDrive, S3, Web Crawler, Granola)\n- Framework Integrations (LangChain, CrewAI, OpenAI SDK, Vercel AI SDK, LangGraph, 15+ more)\n- SuperRAG (managed retrieval-augmented generation)\n- SMFS (Semantic Memory File System)\n- MCP (Model Context Protocol integration)\n- MemoryBench (open-source benchmarking framework)\n- Migration Guides (from Mem0 and Zep)\n- API Reference (connections, container tags, content, documents, ingestion, profiles, search, settings)\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-supermemory-research",
      "to": "page:docs-supermemory-research-raw-01-docs-index",
      "kind": "contains_page"
    }
  ]
}