Add document

import Supermemory from "supermemory"; const client = new Supermemory({ apiKey: process.env.SUPERMEMORY_API_KEY, }); const doc = await client.add({ content: "Sarah is being promoted to VP of Product.", containerTag: "user_123", customId: "chat_session_1", metadata: { type: "conversation" }, dreaming: "instant", }); console.log(doc.id, doc.status); // queued → poll until done

Authorizations

Authorization

string

header

required

Bearer authentication header of the form Bearer <token>, where <token> is your auth token.

Body

application/json

content

string

required

The content to extract and process into a document. This can be a URL to a website, a PDF, an image, or a video.

containerTag

string

Optional tag this document should be containerized by. Max 100 characters, alphanumeric with hyphens, underscores, and dots only.

containerTags

string[]

deprecated

entityContext

string

Optional entity context for this container tag. Max 1500 characters. Used during document processing to guide memory extraction.

Maximum string length: 1500

customId

string

Optional custom ID of the document. Max 100 characters, alphanumeric with hyphens, underscores, and dots only.

metadata

object

Optional metadata for the document.

Show child attributes

taskType

enum<string>

Task type: "memory" (default) for full context layer with SuperRAG built in, "superrag" for managed RAG as a service.

Available options:

memory,

superrag

filepath

string

Optional file path for the document. Used by supermemoryfs to store the full path of the file.

filterByMetadata

object

Optional metadata filter to apply when pulling related memories and profile during ingestion. Only memories matching these filters will be used as context.

Show child attributes

documentDate

string

When this document's content is from, as opposed to when it was uploaded. Accepts YYYY-MM-DD or a full ISO 8601 timestamp. Memory extraction resolves relative dates against this instead of the ingestion time, and documents in a batch are processed oldest-first so newer facts correctly supersede older ones. Set this whenever you backfill historical content.

dreaming

enum<string>

Processing mode. "dynamic" (default) groups related documents together so memories form from coherent, logical units rather than one isolated entry at a time. "instant" processes each document on its own right away, and bills one extra operation per document.

Available options:

instant,

dynamic

Response

Document added successfully

string

required

Unique identifier of the document

status

string

required

Status of the document

Ingest

Recall

Documents

Memories

Container tags

Connections

Settings

Authorizations

Body

Response