Integration

Long-term memory for Gemini CLI

Google's Gemini CLI is fast, capable, and stateless. Every `gemini -p "..."` call starts fresh, with no memory of yesterday's question, last week's project context, or your standing preferences. Add Engram to `~/.gemini/settings.json` and Gemini gets six durable memory tools without changing your CLI muscle memory.

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 Gemini CLI.

Three steps to memory in your agent

Sign up. Free, no card. You'll land on a Getting Started page that walks the next two steps.
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.
Paste the snippet below into your agent and restart it. Use Authorization: Bearer <api-key>with the API key from your portal.

engram-gemini-cli: Google's official Gemini CLI

Adds six Engram MCP tools to gemini via ~/.gemini/settings.json. The url field is the signal; Gemini auto-detects SSE. Source: github.com/lumetra-io/engram-gemini-cli.

Create or merge into ~/.gemini/settings.json (or .gemini/settings.json for project scope):

~/.gemini/settings.json

{
  "mcpServers": {
    "engram": {
      "url": "https://mcp.lumetra.io/mcp/sse",
      "headers": {
        "Authorization": "Bearer <api-key>"
      },
      "timeout": 30000
    }
  }
}

Verify:

Terminal

npm install -g @google/gemini-cli
gemini -p "Use the engram store_memory tool to remember 'gemini cli verification'. Then call engram list_memories on bucket default with limit 3."

What you can do once memory's wired in

Recall the answers Gemini gave you on the same topic last week without retyping the context
Remember which information sources you trust for which topic (finance, medical, code, travel)
Pull up the spec for that side project you haven't touched in a month
Avoid re-explaining your team, org, and role every time you ask Gemini a work-context question

FAQ

Does Gemini CLI auto-detect SSE vs stdio?

Yes. Provide a `url` field and Gemini connects over SSE. Provide `command` and `args` for stdio servers. The presence of `url` is the signal.

Can I scope the config per project?

Yes. A `.gemini/settings.json` in the project root overrides `~/.gemini/settings.json` for that project. Useful for project-specific buckets.

Does this require a paid Gemini API tier?

Engram works on any Gemini CLI install; your LLM credentials are unrelated. Engram is BYOK for the *extraction + synthesis* layer (configured at lumetra.io/models), not for Gemini's own model calls.

Related integrations

Ship durable memory in Gemini CLI 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