{
  "id": "page:docs-supermemory-research-raw-05-api-reference",
  "_kind": "Page",
  "_file": "wiki/docs/supermemory-research/raw/05-api-reference.md",
  "_cluster": "wiki",
  "attributes": {
    "nodeKind": "Page",
    "sourcePath": "docs/supermemory-research/raw/05-api-reference.md",
    "sourceKind": "repo-docs",
    "title": "Supermemory API Reference",
    "displayName": "Supermemory API Reference",
    "slug": "docs/supermemory-research/raw/05-api-reference",
    "articlePath": "wiki/docs/supermemory-research/raw/05-api-reference.md",
    "article": "\n# Supermemory API Reference\n\nSource: https://supermemory.ai/docs/api-reference\n\n## Base URL\n\n`https://api.supermemory.ai`\n\n## Authentication\n\nBearer token (HTTP bearer scheme). API keys prefixed with `sm_`.\n\n## POST /v3/documents\n\n### Purpose\n\nIngest documents with flexible content types (text, URLs, files, images, videos) and optional metadata.\n\n### Request Body\n\n**Required:**\n- `content` (string): The content to process -- can be a website URL, PDF, image, or video\n\n**Optional Parameters:**\n- `containerTag`: Organize documents with a tag (max 100 chars; alphanumeric with hyphens, underscores, dots)\n- `entityContext`: Guide memory extraction (max 1,500 chars)\n- `customId`: Custom document identifier (max 100 chars)\n- `metadata`: Key-value pairs (strings, numbers, booleans, or string arrays)\n- `taskType`: Choose \"memory\" (full context layer, default) or \"superrag\" (managed RAG service)\n- `filepath`: File path information for filesystem integration\n- `filterByMetadata`: Filter related memories by metadata criteria during ingestion\n- `dreaming`: Processing mode -- \"dynamic\" (default, batches related documents) or \"instant\" (processes immediately)\n\n### Response\n\n**Success (200):**\n```json\n{\n  \"id\": \"string\",\n  \"status\": \"string\"\n}\n```\n\n**Errors:**\n- 401: Unauthorized (missing/invalid bearer token)\n- 500: Internal server error\n\n## Additional API Operations\n\n### Memory Operations\n- Create memories directly\n- Update and version memory entries\n- Soft-delete memories (\"forgetting\")\n- Search across indexed content\n\n### Document Operations\n- Add single or batch documents\n- Upload files for processing\n- Update document metadata\n- Delete by ID or custom identifiers\n- Track processing status\n- Access document chunks and presigned URLs\n\n### Profile Operations\n- Retrieve user profiles\n- Get static + dynamic context\n- Query with threshold filtering\n\n### Search Operations\n- Semantic search with metadata filtering\n- Hybrid search (RAG + memory)\n- Low-latency retrieval for conversational AI\n\n## SDK Usage\n\n### JavaScript/TypeScript\n```javascript\nimport { Supermemory } from \"supermemory\";\nconst client = new Supermemory({ apiKey: \"sm_...\" });\n\n// Add memory\nawait client.add({\n  content: \"User prefers functional patterns\",\n  containerTag: \"user_123\",\n});\n\n// Retrieve profile\nconst { profile, searchResults } = await client.profile({\n  containerTag: \"user_123\",\n  q: \"programming style?\",\n});\n\n// Hybrid search\nconst results = await client.search.memories({\n  q: \"how to deploy?\",\n  containerTag: \"user_123\",\n  searchMode: \"hybrid\",\n});\n```\n\n### Python\n```python\nfrom supermemory import Supermemory\nclient = Supermemory(api_key=\"sm_...\")\n\nclient.add(content=\"User prefers functional patterns\", container_tag=\"user_123\")\nprofile = client.profile(container_tag=\"user_123\", q=\"programming style?\")\n```\n",
    "documents": []
  },
  "outgoingEdges": [],
  "incomingEdges": [
    {
      "from": "page:docs-supermemory-research",
      "to": "page:docs-supermemory-research-raw-05-api-reference",
      "kind": "contains_page"
    }
  ]
}