Mimirs
@TheWinci
About Mimirs
Persistent project memory for AI coding agents. Hybrid vector + BM25 search with AST-aware chunking (24 languages), dependency graph boosting, conversation memory, checkpoints, annotations, and wiki generation. Local-first, zero config. `bunx mimirs init`
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"mimirs": {
"command": "bunx",
"args": [
"mimirs@latest",
"serve"
],
"env": {
"RAG_PROJECT_DIR": "path/to/your/folder"
}
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Mimirs?
Mimirs is a persistent project memory system for AI coding agents. It provides hybrid vector and BM25 search with AST-aware chunking for 24 programming languages, dependency graph boosting, conversation memory, checkpoints, annotations, and wiki generation. It is local-first and requires zero configuration.
How to use Mimirs?
Initialize a Mimirs project by running bunx mimirs init in your project directory.
Key features of Mimirs
- Hybrid vector + BM25 search
- AST-aware chunking for 24 languages
- Dependency graph boosting
- Conversation memory, checkpoints, and annotations
- Wiki generation
- Local-first, zero configuration
Use cases of Mimirs
- Provide AI coding agents with persistent project-level memory
- Enable context-aware search across a codebase
- Maintain conversation history and checkpoints during development
- Auto-generate project wiki from code structure
FAQ from Mimirs
What does Mimirs do that alternatives do not?
Mimirs combines vector search with BM25, AST-aware chunking for 24 languages, dependency graph boosting, conversation memory, checkpoints, annotations, and wiki generation in a single local-first tool.
What are the runtime requirements?
Mimirs requires only bun (to run bunx mimirs init) and runs locally with no external services.
Where does Mimirs store its data?
All data is stored locally on the user’s machine; Mimirs is local-first.
What transport or authentication does Mimirs use?
The README does not specify transport or authentication mechanisms.
Are there any known limitations?
The README does not list any known limitations.
Frequently asked questions
What does Mimirs do that alternatives do not?
Mimirs combines vector search with BM25, AST-aware chunking for 24 languages, dependency graph boosting, conversation memory, checkpoints, annotations, and wiki generation in a single local-first tool.
What are the runtime requirements?
Mimirs requires only `bun` (to run `bunx mimirs init`) and runs locally with no external services.
Where does Mimirs store its data?
All data is stored locally on the user’s machine; Mimirs is local-first.
What transport or authentication does Mimirs use?
The README does not specify transport or authentication mechanisms.
Are there any known limitations?
The README does not list any known limitations.
Basic information
More Memory & Knowledge MCP servers
Zettelkasten MCP Server
entanglrA Model Context Protocol (MCP) server that implements the Zettelkasten knowledge management methodology, allowing you to create, link, explore and synthesize atomic notes through Claude and other MCP-compatible clients.
Solomd
zhitongblogA markdown editor — and the bridge to your LLM. Local-first, MIT, ~15 MB. Bundled MCP server lets Claude Code / Codex / Cursor drive your vault directly. 14 AI providers BYOK.
Obsidian MCP Server
StevenStavrakisA simple MCP server for Obsidian
MCP Apple Notes
RafalWilinskiTalk with your notes in Claude. RAG over your Apple Notes using Model Context Protocol.
Context Portal MCP (ConPort)
GreatScottyMacContext Portal (ConPort): A memory bank MCP server building a project-specific knowledge graph to supercharge AI assistants. Enables powerful Retrieval Augmented Generation (RAG) for context-aware development in your IDE.
Comments