Integration
Long-term memory for Continue
Continue runs in VS Code, JetBrains, and the `cn` CLI, and each surface forgets independently. Add Engram via a single YAML file in `~/.continue/mcpServers/` and every Continue surface, on every machine signed into your account, shares one durable memory.
Install
Three steps: sign up for an Engram API key, paste a BYOK LLM-provider key on /models, then drop the snippet below into Continue.
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-continue-dev: Continue (VS Code, JetBrains, cn CLI)
Native SSE config in ~/.continue/mcpServers/engram.yaml. Note: MCP tools only fire in Continue's Agent mode, not plain chat. Source: github.com/lumetra-io/engram-continue-dev.
- Create
~/.continue/mcpServers/engram.yaml(user-level) or<project>/.continue/mcpServers/engram.yaml(project-scoped): - Restart Continue and switch to Agent mode; the six Engram tools appear in the tool palette.
name: engram
version: 0.0.1
schema: v1
mcpServers:
- name: engram
type: sse
url: https://mcp.lumetra.io/mcp/sse
requestOptions:
headers:
Authorization: Bearer <api-key>What you can do once memory's wired in
- Recall the build / test / lint conventions you've configured for each language across your projects
- Remember the API patterns you prefer when scaffolding new services (REST vs GraphQL, auth shape, error envelope)
- Pull up debugging sessions where we found the root cause of a flaky test
- Capture team-level decisions so they propagate from VS Code to JetBrains seamlessly
FAQ
MCP tools aren't firing in Continue. Why?
Continue only invokes MCP tools in Agent mode, not plain chat. Switch via the sidebar mode picker. The `cn` CLI exposes MCP tools by default in its agent-style invocation, so no mode flag is needed there.
Can I use a different config per project?
Yes. Put `engram.yaml` under `<project>/.continue/mcpServers/` to scope it. For configuration that follows you across projects, put it in your global `~/.continue/config.yaml` under `mcpServers` (the officially documented global path).
Is `${{ secrets.X }}` syntax supported for the Authorization header?
Continue documents `${{ secrets.X }}` for env values bound to MCP servers; headers in `requestOptions` are documented as static today. Put the YAML under your user-level dir (not in a project repo) and treat it as a secret.
Related integrations
Ship durable memory in Continue 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.