Octobrain
Give your coding agent durable recall across sessions so decisions, preferences, and project facts survive context resets.
Overview
io.github.Muvon/octobrain is an MCP server for the Build phase that gives AI assistants persistent memory with semantic search and knowledge-graph relationships.
What is this MCP server?
- Persistent memory store across chats and repos
- Semantic search over stored facts and notes
- Knowledge-graph style relationships between entities
- MCP server integration for Claude Code, Cursor, and other agents
- Version 0.3.0 Model Context Protocol server package
- Server version 0.3.0 per publisher server.json
- Described as persistent memory with semantic search and knowledge graph relationships
What problem does it solve?
Agent sessions forget your product context, so you waste tokens re-briefing the model on the same facts every time you open a new chat.
Who is it for?
Indie builders running long Claude Code or Cursor projects who need cross-session recall of specs, stack choices, and customer notes.
Skip if: Teams that already use a dedicated knowledge base with tight ACLs and do not want the agent to maintain its own memory graph.
What do I get? / Deliverables
After registration, your agent can query stored memory semantically and relate entities so multi-day solo builds stay coherent without manual paste-ins.
- Queryable persistent memory accessible from agent tool calls
- Semantic retrieval over stored assistant knowledge
- Linked entities for multi-fact project context
Recommended MCP Servers
Journey fit
Persistent assistant memory is cataloged under Build because it is installed as agent infrastructure, even though you benefit from it in every phase of shipping. Agent-tooling is the canonical shelf for MCP servers that extend what the assistant can remember and retrieve, not a one-off app feature.
How it compares
MCP memory server, not a single-repo RAG indexer like octocode.
Common Questions / FAQ
Who is io.github.Muvon/octobrain for?
Solo and small-team builders who use MCP-capable agents and want semantic, persistent memory beyond the current chat window.
When should I use io.github.Muvon/octobrain?
Use it when you iterate on one product over weeks and need the agent to remember decisions, entities, and relationships without repeating full briefs.
How do I add io.github.Muvon/octobrain to my agent?
Add the octobrain MCP server entry to your client’s MCP config (stdio or hosted transport per publisher docs), restart the agent, and grant tool access for memory read/write tools.