Login

Integration

Long-term memory for Claude.ai

Claude.ai's Projects let you attach static knowledge, but they don't capture what you and Claude work out together. Add Engram as a custom connector and Claude remembers across every conversation, project, and device. OAuth handles auth, so there's no API key to paste.

Get an API key (free) All integrations →

Install

Three steps: sign up for an Engram API key, paste a BYOK LLM-provider key on /models, then drop the snippet below into Claude.ai (web).

Three steps to memory in your agent

  1. Sign up. Free, no card. You'll land on a Getting Started page that walks the next two steps.
  2. Add your LLM key. Engram is BYOK. Paste an OpenAI / Anthropic / Groq / Together / Fireworks key and we'll route every extraction and query call through your provider. You pay your provider directly. We never see your inference.
  3. Paste the snippet below into your agent and restart it. Use Authorization: Bearer <api-key>with the API key from your portal.

OAuth: no config file, no key to paste

Claude.ai handles auth via OAuth. Sign in to Lumetra in the popup and you're done.

  1. In Claude.ai, open Settings → Connectors (Pro/Max/Team/Enterprise plans) and click Add custom connector.
  2. Name it Engram and paste this server URL:
    3. MCP server URL
    https://mcp.lumetra.io/mcp/sse
  3. Confirm. Claude.ai opens a popup to lumetra.io. Sign in (or create a free account) and approve access. You'll land back in Claude.ai with the connector live.
  4. Start a new chat and verify by asking Claude something like "What memory tools do you have?". It should list store_memory, query_memory, and the rest.

Heads up: BYOK still required

Engram is BYOK end-to-end. After you're connected, paste an OpenAI / Anthropic / Groq / Together / Fireworks key on the Lumetra portal so store_memory / query_memory don't return 412.

What you can do once memory's wired in

  • Remember the writing style guide we worked out across last week's threads
  • Recall the strategic decisions from the planning conversation when you open a new chat
  • Pull up your partner's preferences, allergies, and important dates for any gift-related question
  • Capture facts about your team (names, roles, working hours) once and let Claude reuse them

FAQ

Which Claude.ai plans support custom connectors?

Free, Pro, Max, Team, and Enterprise. Free is capped at one custom connector; paid plans have higher limits. On Team and Enterprise an Owner has to add the connector at the workspace level before individual users can enable it.

Do I need to paste an API key?

No. Claude.ai handles auth via OAuth. You'll be redirected to Lumetra to sign in (or create a free account), then bounced back with the connector live.

Does it work in the Claude.ai mobile apps?

Yes. Connectors are account-scoped, not device-scoped. Add it once on web and mobile picks it up automatically.

Related integrations

ChatGPT (web) OAuth app for ChatGPT (GPT-5 / Thinking). Connector list caches aggressively, so a full walkthrough is included. Read more → Claude Code Native Claude Code plugin: install via the plugin marketplace, with /engram-remember and /engram-recall slash commands. Read more → LibreChat Self-hosted ChatGPT alternative. OAuth per LibreChat user (no shared key in librechat.yaml), with a shared-bearer alternative for single-user deploys. Read more →

Ship durable memory in Claude.ai (web) today

Free tier: 10K memories and 50K retrievals per month. No credit card. Same Engram backend powers all 41 integrations, so memories you write from one client are immediately queryable from the rest.

Start free See pricing