{ integrations }

Memory for the tools
you already use.

Connect models, frameworks, data sources, and productivity tools. MatiAI sits quietly between them and keeps context flowing.

Request an integration
Context workflow
Live
Production pipeline Running
TRIGGERNew message#product
CONTEXTMatiAIExtract and resolve
MODELGenerate reply4 memories injected
OUTPUTUpdate briefNorthstar launch
Last run completed184 msView execution
AllModelsFrameworksToolsDataData sourcesProductivity

Models

OpenAI

Native integration

Anthropic

Connect in a few lines

Google Gemini

Native integration

Mistral

Connect in a few lines

Frameworks

Vercel AI SDK

Native integration

LangChain

Connect in a few lines

Ollama

Native integration

Hugging Face

Connect in a few lines

Tools

MCP

Native integration

Cursor

Connect in a few lines

GitHub

Native integration

Slack

Connect in a few lines

Data

PostgreSQL

Native integration

Notion

Connect in a few lines

Google Drive

Native integration

S3

Connect in a few lines

Data sources

Supabase

Native integration

MongoDB

Connect in a few lines

Redis

Native integration

Google Cloud

Connect in a few lines

Productivity

Linear

Native integration

Figma

Connect in a few lines

Discord

Native integration

GitHub

Connect in a few lines

PER-INTEGRATION

Connect it once. It just works.

Every integration speaks the same memory API — ingest, recall, and sync — so wiring a new tool is a config change, not a project.

OpenAIModel

Stream memories into any OpenAI chat or agent as ranked context, and persist new facts on every turn.

  • Tool calling
  • Streaming
  • Function context
Connect
AnthropicModel

Inject durable context into Claude and recall memories across long-running agent sessions.

  • MCP native
  • Long context
  • Agent memory
Connect
Vercel AI SDKFramework

Add memory to any AI SDK route in a few lines with typed helpers and edge-ready runtime.

  • Edge runtime
  • Typed helpers
  • Stream UI
Connect
LangChainFramework

Use MatiAI as a memory store and retriever inside LangChain graphs and agents.

  • Retriever
  • Agent memory
  • Graph nodes
Connect
MCPTool

Expose the memory engine to Claude Desktop, Cursor, and any MCP-compatible client instantly.

  • One-command
  • Bi-directional
  • Local + remote
Connect
SlackTool

Turn Slack conversations into structured memory and trigger recall on keywords or mentions.

  • Ingestion
  • Recall triggers
  • DM context
Connect
PostgreSQLData

Sync memories into your warehouse, or use Postgres as a self-hosted storage backend.

  • Export
  • Self-host
  • CDC sync
Connect
NotionData

Keep Notion docs in sync with memory so decisions and briefs stay current automatically.

  • Doc sync
  • Two-way
  • Scheduled
Connect

Need an integration we do not have yet?

Build your own in an afternoon with the HTTP API and webhooks — or request one and we will prioritize it.

Request an integration

{ Start building }

Give your AI a memory it can trust.

Free to start. Open by design. Ready for production.

Get MatiAI