Connect Supermemory to Your AI Agent

Store known facts, index documents asynchronously, recall memories or chunks, and correct versioned memory records. Ask your connected AI assistant to explain the results or carry out the specified operation.

8 tools API key

Try a Supermemory Task

Copy a prompt into your agent with the oo Skill installed. Add your page links and source notes before running it.

oo Skill Supermemory

Recall a Tenant’s Context

Search and read profile context within one explicit container.

/oo Supermemory: Search Supermemory container [containerTag] for [question], then get its profile context and summarize returned facts with uncertainty.

Index a Reference Document

Submit content for an explicit task type and inspect processing.

/oo Supermemory: Add Supermemory document content [text or URL] in containerTag [container], taskType [type], then get the returned document ID to check processing.

Start Working with Supermemory

  1. Install OOMOL

    Set up oo CLI and the oo Skill in your agent.

    Install OOMOL →
  2. Connect Supermemory

    Connect with a Supermemory API key.

    Connect Supermemory →
  3. Give Your Agent a Task

    Use /oo with the pages to read, the changes to make, and where to save the result.

    Connection Guide →

Choose Your Agent

  • A Supermemory API key and tenant containerTag.
  • Use containerTag to isolate tenant context. Document indexing is asynchronous, while create_memories stores already-known facts for immediate recall.

Supermemory Tools

Supermemory connects supported service operations to your AI assistant. Store known facts, index documents asynchronously, recall memories or chunks, and correct versioned memory records.

Recall

Search memories or chunks and retrieve profile context.

Memory Writes

Store known facts, index documents, or correct a memory while retaining history.

8 tools

Create Memories

Store one or more already-known facts directly as immediately searchable Supermemory memories.

supermemory.create_memories

Add Document

Submit raw text, a transcript, or a public URL for asynchronous memory extraction or managed RAG indexing.

supermemory.add_document

Get Document

Get a Supermemory document and its current processing status, including the final extracted content when available.

supermemory.get_document

Search

Recall relevant memories, document chunks, or both for a question within one tenant container.

supermemory.search

Get Profile

Retrieve long-term, recent, and bucketed profile context for one user or tenant container.

supermemory.get_profile

Update Memory

Correct one memory by ID, creating a new version while preserving its prior version history.

supermemory.update_memory

Forget Memory

Soft-delete one memory by ID so it is excluded from normal recall while remaining auditable.

supermemory.forget_memory

Delete Document

Permanently delete one source document by its Supermemory ID or caller-provided custom ID.

supermemory.delete_document

Frequently Asked Questions

How do I separate tenants?

Use a consistent containerTag per user or tenant for writes and recall.

Is document indexing immediate?

No. Read get_document processing status. create_memories is for already-known immediately searchable facts.

Does correction erase earlier memory versions?

update_memory creates a new version while preserving prior history.

Connect Supermemory

About Supermemory ↗