Mcp Server
Give Claude Code or Cursor a persistent memory layer so coding sessions remember decisions, preferences, and past context via semantic search.
Overview
in.creo-memories/mcp-server is a MCP server for the Build phase that lets AI agents store and recall project memories through semantic search.
What is this MCP server?
- Remote streamable-http MCP at mcp.creo-memories.in (no local server required)
- Semantic search over stored memories for fuzzy recall instead of exact keyword match
- Designed for AI agents that need durable context across sessions and projects
- Open-source creo-memories repo on GitHub for self-host or fork paths
- Version 2.0.0 in the published MCP server manifest
- Server manifest version 2.0.0
- Single remote endpoint: mcp.creo-memories.in (streamable-http)
- Source repository: github.com/chronista-club/creo-memories
What problem does it solve?
Coding agents forget everything between sessions, so you waste time re-stating context the agent should already know.
Who is it for?
Solo builders running long-lived agent workflows who want hosted memory without operating their own embedding stack.
Skip if: Teams that need enterprise ACLs, audit trails, or memory that must stay strictly on-prem with no remote endpoint.
What do I get? / Deliverables
After you register the remote MCP, your agent can persist and semantically retrieve memories so multi-day solo builds stay coherent.
- Registered remote MCP connection in your agent
- Ability to store and semantically recall agent memories during coding tasks
Recommended MCP Servers
Journey fit
Agent memory MCPs belong on the build shelf under agent-tooling because they extend how your coding agent works day to day, even though recall helps across later operate and grow work. Semantic store-and-recall is core agent infrastructure, not a one-off integration for a single API endpoint.
How it compares
MCP memory backend for agents, not a brainstorming or planning skill.
Common Questions / FAQ
Who is in.creo-memories/mcp-server for?
It is for indie developers and small teams using MCP-enabled coding agents who need durable, searchable memory across sessions.
When should I use in.creo-memories/mcp-server?
Use it once you start repeating the same context to your agent or when multiple features share decisions you want recalled automatically.
How do I add in.creo-memories/mcp-server to my agent?
Add the remote MCP URL https://mcp.creo-memories.in with type streamable-http in Claude Code, Cursor, or another MCP client, then verify tools appear in the agent panel.