{
  "markdown": "# Krimto\n\n> Open-source memory for AI coding agents you own in git — governed by a real user→team→org hierarchy, not a vendor database. Solo to team. Apache-2.0.\n\n[![License: Apache 2.0](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)\n[![npm](https://img.shields.io/npm/v/@krimto-labs/krimto)](https://www.npmjs.com/package/@krimto-labs/krimto)\n[![CI](https://github.com/krimto-labs/krimto/actions/workflows/test.yml/badge.svg)](https://github.com/krimto-labs/krimto/actions/workflows/test.yml)\n\nYour agent's memory, as markdown in your own git — synced across every editor and machine. Tell your\nagent \"remember X\" in any editor and it saves a durable, attributable fact; ask later — in a new chat,\na different editor, or on another machine — and it recalls the right answer. One command, no account,\nno vendor database: your memory follows you.\n\nYou own it: every fact is a plain markdown file in a git repo you control — readable, reviewable in a\nPR, and yours. When a teammate joins, that same memory becomes a shared team brain governed by a real\n`user → team → org` hierarchy (the most specific scope wins at recall) — the one primitive no other\nmemory tool ships. Solo is free; the governance is what you grow into.\n\n## See it\n\nOne `npx` command sets it up; then your agent saves a fact in Claude Code and recalls it in Cursor —\nno account, no database:\n\n<!-- media/krimto-demo.gif is the lead demo (docs/ is gitignored, so the README can't link there).\n     The full three-demo storyboard lives at docs/demos/krimto-demos.html (local). -->\n![Krimto — save a fact in Claude Code, recall it in Cursor](media/krimto-demo.gif)\n\nThree short demos — **solo** (remember in Claude Code, recall in Cursor), **cross-machine** (write,\n`git push`, recall on another laptop), and **team** (a teammate's fact reaches you; your personal notes\nstay yours).\n\nBecause every memory is just a file in your git, you can read it yourself — no API, no dashboard needed:\n\n```console\n$ cat ~/.krimto/user/you@example.com/favorite-color.md\n---\nid: fct_01KT3BDSYY1KK80SG3S7KBEV58\nscope: user/you@example.com\ntitle: Favorite color\nauthor: you@example.com\ncreated: 2026-06-02T05:03:14Z\nupdated: 2026-06-02T05:03:14Z\ntags:\n  - personal\nsource: claude-code\n---\n\nUser's favorite color is red.\n\n$ git -C ~/.krimto log --oneline -1\n8bc5b86 krimto: write batch — 1 fact\n```\n\nThat's the whole bet: **your agent's memory is plain markdown in a git repo you own — not rows in\nsomeone else's database.**\n\n## Why Krimto\n\n- **You own it, in your git.** Every fact is a markdown file in a git repo you control — `git log` the\n  audit trail, edit it in any editor, review it in a pull request. No vendor database, no lock-in.\n- **Governed `user → team → org` hierarchy.** Knowledge is scoped to a person, a team, or the whole\n  company, and the most specific scope wins at recall — a server-enforced primitive no other memory\n  tool ships for free.\n- **Follows you everywhere — cross-vendor, Apache-2.0.** One MCP server syncs the same memory across\n  Claude Code, Cursor, Codex, Gemini CLI and more — every editor and machine. Fully open source, with\n  no managed-service restriction.\n\n## Try it in 2 minutes (solo, no account)\n\n```bash\nnpx @krimto-labs/krimto init\n```\n\nThe setup wizard detects your editor, wires it up, and turns on automatic memory. Then, in any chat:\n\n```\n\"Remember that our staging DB resets every Sunday.\"\n```\n\nOpen a new chat and ask:\n\n```\n\"What do you know about staging?\"   → it remembers.\n```\n\nSee your notes with `krimto notes` (terminal) or `krimto ui` (browser dashboard). Your data lives in\n`~/.krimto` — the same folder no matter which project you're working in.\n\n## Connect your agent\n\n`krimto init` wires supported editors for you. What auto-connects vs. needs one copy-paste step:\n\n| Editor | Setup |\n|---|---|\n| Cursor | auto-connects |\n| Claude Code | auto-connects |\n| Codex | manual snippet |\n| Gemini CLI | manual snippet |\n\nTo connect any MCP client manually, point it at Krimto over stdio:\n\n```bash\nclaude mcp add krimto -- npx -y @krimto-labs/krimto\n```\n\n…or the config-file form (Cursor, Codex, Gemini CLI, etc. use the same shape):\n\n```json\n{ \"mcpServers\": { \"krimto\": { \"command\": \"npx\", \"args\": [\"-y\", \"@krimto-labs/krimto\"] } } }\n```\n\nBy default an agent uses Krimto only when you ask. Running `krimto init` once in your project drops a\nstanding rule so it uses Krimto on its own.\n\n### Install as a Claude Code plugin\n\nPrefer Claude Code's plugin system? Add Krimto's marketplace and install it directly:\n\n```bash\n/plugin marketplace add krimto-labs/krimto\n/plugin install krimto@krimto\n```\n\nThis bundles the MCP server together with Krimto's skills, the `/krimto-status` command, and the\nmemory hooks — no separate `krimto init` needed.\n\n## How it works\n\nThree layers, one source of truth:\n\n1. **Storage** — facts are markdown files in a git repository (the source of truth; git is the audit log).\n2. **Index** — a SQLite + sqlite-vec hybrid index (keyword + vector) for fast retrieval, with scope\n   precedence applied at ranking time.\n3. **Access** — an API server enforces who can read and write each scope (`user` / `team` / `org`).\n\n## Team mode\n\nWhen you're ready to share memory with teammates:\n\n```bash\nnpx @krimto-labs/krimto team init\n```\n\nThis walks you through an admin email, your org/team name, an optional shared git remote, and\nteammate invites — then prints a join command for each teammate:\n\n```bash\nkrimto join --server <url> --key <key>\n```\n\nTeammates can connect to one shared server, or each run their own Krimto synced over a shared git\nremote. Personal and team notes live together and sync as a unit. Step back to solo any time with\n`krimto team disband` (your notes are preserved).\n\n## Self-host\n\nKrimto runs anywhere Node 20+ runs.\n\n```bash\n# HTTP server + browser dashboard at http://localhost:8080\nnpx @krimto-labs/krimto serve\n\n# or Docker\ndocker run -d -p 8080:8080 -v ~/.krimto:/data ghcr.io/krimto-labs/krimto:latest\n```\n\nRun `npx @krimto-labs/krimto --help` for the full command surface.\n\n## Roadmap\n\nv0.2 (current) ships the memory core, teams, the web dashboard, and the cross-vendor MCP server.\nNext: OAuth sign-in and a pull-request approval flow (v0.3), then a hosted Krimto Cloud (v1.0). See\n[ROADMAP.md](ROADMAP.md).\n\n## Contributing & license\n\nContributions welcome — see [CONTRIBUTING.md](CONTRIBUTING.md) and our\n[Code of Conduct](CODE_OF_CONDUCT.md). Security reports: [SECURITY.md](SECURITY.md).\n\nLicensed under [Apache-2.0](LICENSE). The same code is self-hostable by a solo developer or an\nenterprise — no tier walls.\n",
  "bytes": 6601,
  "sha": "8092cd7577a3b366d7f3b01fcd7fd51eba7bd989592a76b5d32edf281555f21f",
  "repo_slug": "krimto-labs/krimto",
  "fonte": "repo",
  "truncated": false,
  "api": "https://agentalog.com/api/listings/mcp_io_github_krimto_labs_krimto_c1608b6e/readme"
}