Lore
Persist and recall agent memory across Claude Code, Cursor, and other MCP sessions so long-running solo projects do not lose context every chat reset.
Overview
Lore is a Build-phase MCP server that provides lightweight persistent memory so coding agents can retain context across sessions and clients.
What is this MCP server?
- Lightweight MCP server positioned for cross-session, cross-agent memory
- npm package @jordanhindo/lore at version 0.1.1
- stdio transport for standard local MCP registration
- Designed for any agent, any session framing per upstream description
- Complements stateless LLM chats with an external memory layer
- npm package @jordanhindo/lore version 0.1.1
- stdio transport per server.schema.json
- Repository source github.com/jordanhindo/lore
Community signal: 9 GitHub stars.
What problem does it solve?
Every new agent session forces solo builders to re-explain stack choices, naming rules, and open tasks because chat context does not persist by default.
Who is it for?
Indie developers running multi-week agent-assisted builds who swap between MCP clients and need a shared memory substrate.
Skip if: Teams that already centralize knowledge in a strict compliance wiki with no agent write path, or projects needing only single-session codegen.
What do I get? / Deliverables
After registering Lore, your agents can read and write durable memory through MCP instead of relying only on the current thread history.
- Registered Lore MCP server in your agent configuration
- Reusable memory reads and writes across sessions
- Reduced re-onboarding time when starting new chats on the same project
Recommended MCP Servers
Journey fit
Durable agent memory is build-phase agent tooling—you install it while shaping how your coding agent works, and it keeps paying off through ship and operate. Agent-tooling is canonical because Lore is infrastructure for the agent itself, not application UI or backend business logic.
How it compares
Agent memory MCP layer, not a vector database admin console or a project-management SaaS integration.
Common Questions / FAQ
Who is Lore for?
Lore is for solo builders and agent-first developers who want persistent recall across Claude Code, Cursor, and other MCP sessions without rebuilding context each time.
When should I use Lore?
Use it while setting up agent tooling in the build phase and whenever long-running work spans many chats, handoffs, or different MCP clients on the same codebase.
How do I add Lore to my agent?
Install @jordanhindo/lore from npm, add it as an stdio MCP server in your client config, then invoke its memory tools from your agent according to the Lore repository documentation.