Krimto
@krimto-labs
About Krimto
Open-source memory for AI coding agents. Tell your agent to "remember" something in Cursor, Claude Code, Codex, or Gemini CLI, and Krimto saves it as a plain markdown file in a git repo you own — not rows in a vendor's database. Recall it later in a new chat, a different editor,
Basic information
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"krimto": {
"command": "npx",
"args": [
"-y",
"@krimto-labs/krimto"
]
}
}
}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 Krimto?
Krimto is an open-source memory system for AI coding agents that stores facts as plain markdown files in a git repository you own. It enforces a user→team→org hierarchy for scoping knowledge and syncs across editors and machines without any vendor database. It is licensed under Apache-2.0 and designed for solo developers and teams.
How to use Krimto?
Install Krimto with npx @krimto-labs/krimto init — the setup wizard detects your editor and enables automatic memory. In any supported editor (Claude Code, Cursor, Codex, Gemini CLI), tell your agent “remember X” to save a fact; later ask “what do you know about X?” to recall it. Run krimto notes in the terminal or krimto ui for a browser dashboard. For team memory, use npx @krimto-labs/krimto team init and share the join command with teammates.
Key features of Krimto
- Facts are plain markdown files in a git repo you control.
- User → team → org hierarchy governs who sees what.
- Cross-editor and cross-machine sync via one MCP server.
- Open source under Apache-2.0 license.
- SQLite + vector index for fast keyword and semantic recall.
- Self-hostable via
npxor Docker on any Node 20+ system.
Use cases of Krimto
- Solo developer: remember project context across multiple editors and machines.
- Team collaboration: share relevant team knowledge while keeping personal notes private.
- Compliance and audit: review every fact change through git history and pull requests.
- Self-hosted enterprise: run Krimto behind your own firewall without vendor lock‑in.
FAQ from Krimto
What runtime does Krimto require?
Krimto runs anywhere Node 20+ is available. You can also run it in Docker with docker run -d -p 8080:8080 -v ~/.krimto:/data ghcr.io/krimto-labs/krimto:latest.
Where does Krimto store its data?
All data lives in ~/.krimto — the same folder regardless of which project you are working in. Every fact is a markdown file inside a git repository at that path.
How does team mode work in Krimto?
Team mode creates a shared git remote or a shared Krimto server. Each teammate joins with a command like krimto join --server <url> --key <key>. Personal and team notes sync together, with the most specific scope winning at recall.
Can I use Krimto without creating an account?
Yes, solo mode requires no account. Run npx @krimto-labs/krimto init and you’re ready to go. An account is only needed for the optional hosted Krimto Cloud (planned for v1.0).
How do I connect Krimto to an MCP client manually?
Point any MCP client at Krimto over stdio using: claude mcp add krimto -- npx -y @krimto-labs/krimto or add the equivalent JSON configuration: { "mcpServers": { "krimto": { "command": "npx", "args": ["-y", "@krimto-labs/krimto"] } } }.
More AI & Agents MCP servers
MCP Client for Ollama (ollmcp)
joniglHarness the power of local LLMs with this TUI MCP Client for Ollama. Featuring all core MCP primitives (tools, prompts, resources), agent mode, multi-server, model switching, streaming responses, human-in-the-loop, thinking mode, model params config, system prompts, and saved pre
meGPT - upload an author's content into an LLM
adriancoCode to process many kinds of content by an author into an MCP server
Just Prompt - A lightweight MCP server for LLM providers
dislerjust-prompt is an MCP server that provides a unified interface to top LLM providers (OpenAI, Anthropic, Google Gemini, Groq, DeepSeek, and Ollama)
Shell and Coding agent for Claude and other mcp clients
rusiaamanShell and coding agent on mcp clients
Sequential Thinking Multi-Agent System (MAS)
FradSerAn advanced sequential thinking process using a Multi-Agent System (MAS) built with the Agno framework and served via MCP.
Comments