Memori Mcp
Give Claude Code or Cursor durable recall across sessions so stack choices, user prefs, and project facts survive new chats.
Overview
Memori MCP is a MCP server for the Build phase that adds Memori Cloud persistent recall and augmentation tools to your coding agent.
What is this MCP server?
- stdio MCP package @memorilabs/memori-mcp (npm v1.0.0)
- Recall and augmentation tools backed by Memori Cloud
- Requires MEMORI_API_KEY (secret) from memorilabs.ai
- Persists context so agents do not re-ask the same project questions every session
- Server version 1.0.0; npm identifier @memorilabs/memori-mcp; stdio transport; 1 required secret env var (MEMORI_API_KEY)
Community signal: 3 GitHub stars.
What problem does it solve?
Every new chat makes your agent forget stack decisions, preferences, and prior research you already explained.
Who is it for?
Indie builders running long-lived Claude Code projects who want hosted memory without building their own embedding store.
Skip if: Teams that require fully offline, on-prem-only memory with no third-party API.
What do I get? / Deliverables
Your agent can recall and augment stored memories across sessions so you spend less time re-briefing and more time shipping.
- MCP tools for recall and memory augmentation wired to Memori Cloud
- Cross-session memory continuity for project-specific facts
- Reduced re-prompting when restarting agent threads
Recommended MCP Servers
Journey fit
Memory MCPs sit on the builder’s agent toolchain shelf because they wire persistent recall into day-to-day coding workflows. agent-tooling is where solo builders register MCP servers that extend what the assistant remembers between tasks.
How it compares
Cloud memory MCP integration, not a local-only markdown context skill.
Common Questions / FAQ
Who is Memori MCP for?
Solo and indie developers using MCP-capable agents who want Memori Cloud to remember project and user context between conversations.
When should I use Memori MCP?
Use it when you are in active build or iterate work and the agent keeps losing decisions you already locked in earlier sessions.
How do I add Memori MCP to my agent?
Register the stdio server with npm package @memorilabs/memori-mcp and set the required secret environment variable MEMORI_API_KEY from memorilabs.ai in your MCP config.