Kairn

Context-aware knowledge engine for AI assistants.

Status: pre-1.0. In daily use since February 2026, with 722 tests (see Development) and a published LongMemEval-S benchmark. Interfaces may still change between releases until 1.0. Feedback and issues welcome.

Other tools give your AI a memory. Kairn gives it a knowledge graph with intelligent context routing. It knows what to load, when to load it, and how much - so your AI stays focused, not overwhelmed.

pip install kairn-ai
kairn init ~/brain
kairn serve ~/brain

Add it to Claude Code in one line:

claude mcp add kairn -- kairn serve ~/brain

Or install it as a one-click bundle, no Python setup required: download the .mcpb file from the latest release and open it with a bundle-aware app such as Claude Desktop.

For other clients, see Quick Start below. New to Kairn? Jump to First 5 Minutes.

Install routes

The bundle carries no Kairn source of its own. It declares kairn-ai as a dependency and the host resolves it with uv, so a bundle install and a pip install run identical code. Where the database lives is configurable when you install the bundle; it defaults to ~/.kairn and never leaves your machine.

Why Kairn?

Every AI conversation starts from scratch. Previous insights, decisions, and patterns - gone. Existing memory tools store flat key-value pairs that can't represent relationships or surface the right context at the right time.

Kairn is different:

• Context Router + Progressive Disclosure - Automatically loads relevant subgraphs based on keywords, starting with summaries and drilling into details only when needed. No other tool does this.
• Knowledge Graph with FTS5 - Not flat storage. Typed relationships (depends-on, resolves, causes) between nodes with provenance tracking and full-text search across everything.
• Experience Decay + Auto-Promotion - Experiences lose relevance over time (biological decay model). Frequently-accessed experiences auto-promote to permanent knowledge. Your AI naturally forgets what doesn't matter.
• 22 MCP Tools - Works with Claude Desktop, Cursor, VS Code, Windsurf, and any MCP client. Includes kn_judge for 5-verb relationship judgments and kn_doctor for read-only health diagnostics.
• Per-Workspace Isolation - Each workspace is its own isolated SQLite store. JWT auth and role-based access control (owner / maintainer / contributor / reader) ship for team deployments.

Quick Start

Claude Desktop

Add to ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "kairn": {
      "command": "kairn",
      "args": ["serve", "~/brain"]
    }
  }
}

Cursor

Add to .cursor/mcp.json:

{
  "mcpServers": {
    "kairn": {
      "command": "kairn",
      "args": ["serve", "~/brain"],
      "env": {
        "KAIRN_LOG_LEVEL": "WARNING"
      }
    }
  }
}

VS Code

Add to .vscode/mcp.json:

{
  "servers": {
    "kairn": {
      "type": "stdio",
      "command": "kairn",
      "args": ["serve", "~/brain"]
    }
  }
}

Windsurf

Add to ~/.codeium/windsurf/mcp_config.json:

{
  "mcpServers": {
    "kairn": {
      "command": "kairn",
      "args": ["serve", "~/brain"]
    }
  }
}

Restart your editor. Kairn's 22 tools appear in the MCP section.

First 5 Minutes

A guided first run, end to end:

pip install kairn-ai
kairn init ~/brain              # creates the workspace + database

Add the one-liner from above (or your client's Quick Start snippet), then restart the client. Once connected, ask your assistant to remember something:

"Remember that we chose Postgres over SQLite for the analytics service because we needed concurrent writers."

That calls kn_learn under the hood and returns a JSON envelope like this (captured from a real run, via kairn learn, the CLI mirror of the tool):

{"_v": "1.0", "stored_as": "node", "node_id": "002d9c22", "experience_id": "d0710c2f", "type": "decision", "confidence": "high", "namespace": "knowledge", "candidates": []}

Start a new session and ask it to recall the same thing - that calls kn_recall and surfaces what you just stored, no re-explaining required:

{"_v": "1.0", "count": 2, "results": [
  {"source": "node", "id": "002d9c22", "name": "Decision: we chose Postgres over SQLite for the analytics service beca", "type": "learned_decision", "description": "we chose Postgres over SQLite for the analytics service because we needed concurrent writers", "relevance": 1.0, "relevance_kind": "match"},
  {"source": "experience", "id": "d0710c2f", "type": "decision", "content": "we chose Postgres over SQLite for the analytics service because we needed concurrent writers", "confidence": "high", "relevance": 1.0, "relevance_kind": "recency"}
]}

kn_learn stored both a permanent graph node and a decaying experience (high confidence does both, see Confidence routing); kn_recall found both from a three-word topic.

Read relevance_kind before you read relevance. Both rows above show 1.0 and they do not mean the same thing. match is lexical match strength (bm25); the experience's recency is time-decay - it is 1.0 because the row was created seconds ago, not because it matched well. A third value, similarity, is embedding cosine on the semantic-recall path, and unscored marks a row the surface had no ranking for and filled in with a constant. The numbers are not comparable across kinds, so do not sort a mixed result set on relevance alone. Same caution for min_relevance on kn_recall: it gates nodes on match strength and experiences on recency, one number against two scales. On kn_memories and kn_prune, which see experiences only, it is recency - and on kn_prune it deletes.

Run kairn status ~/brain any time as a smoke test - if it prints a JSON stats block (nodes/edges/experiences counts), the workspace is healthy. Want a scripted tour of every core feature instead of doing it by hand? Run kairn demo ~/brain - it walks through node creation, querying, experience saving, learning, recall, and context in about 30 seconds.

Which tool when

22 tools is a lot to hold in your head on day one. Most sessions only need these:

Everything else (kn_crossref, kn_related, kn_connect, kn_judge, kn_project/kn_projects/kn_log, kn_idea/kn_ideas, kn_promote_pending, kn_prune, kn_remove, kn_status, kn_doctor) is advanced usage - see the full 22 Tools reference below once you're past the basics.

22 Tools (kn_ prefix)

All tools follow MCP protocol with JSON responses.

Graph (6)

Project Memory (3)

Experience Memory (5)

Ideas (2)

Intelligence (5)

Diagnostic (1)

Resources & Prompts

Resources (read-only context for MCP clients):

• kn://status - Graph overview, active project
• kn://projects - All projects with recent progress
• kn://memories - Recent high-relevance experiences

Prompts (session management):

• kn_bootup - Load active project, recent progress, and top memories (session start)
• kn_review - Summarize session and suggest next steps (session end)

How It Works

Architecture

Any MCP Client (Claude, Cursor, VS Code)
        │
        ▼ MCP Protocol (stdio)
FastMCP Server (22 tools)
        │
   ┌────┼────┐
   ▼    ▼    ▼
Graph  Memory  Intelligence
Engine Engine  Layer
   │    │      │
   └────┼──────┘
        ▼
   SQLite + FTS5
   (per-workspace)

Decay Model

Experiences decrease in relevance exponentially:

relevance(t) = initial_score × e^(-decay_rate × days)

Half-lives are calibrated against the real access tail of a production experience store, not guessed (one exception: preference is a new type with no access history yet, so its value is a documented initial estimate until real data accumulates).

Confidence routing via kn_learn:

• high → Permanent node + experience (no decay)
• medium → Experience with 2× decay
• low → Experience with 4× decay
• Auto-promotion: 5+ accesses → permanent node
• Node access tracking: kn_recall, kn_context, and kn_crossref log which nodes were accessed, feeding the decay and promotion pipeline

Benchmarks

Kairn scores 56.2% overall on LongMemEval-S (500/500 questions scored, GPT-4o reader + judge, single run, 0 errors). These are the real per-category numbers, including the bad ones - each red cell links to its diagnosis:

The 500 questions include 30 abstention variants (the right answer is to decline); they are counted inside their categories above and scored separately: Kairn declines correctly on 96.7% of them.

Recall latency is ~1.4 ms per query (FTS5, in-process, no network). Protocol, honesty notes, and reproduction steps: BENCHMARKS.md.

This scorecard stays current: every release that touches recall re-publishes these numbers, and a weak cell stays on the board until the number actually moves. No cherry-picked runs, no hidden categories.

CLI

kairn init <path>              # Initialize workspace
kairn serve <path>             # Start MCP server (stdio)
kairn status <path>            # Graph stats
kairn demo <path>              # Interactive tutorial
kairn benchmark <path>         # Local performance benchmarks (latency, not LongMemEval)
kairn token-audit <path>       # Audit tool token usage
kairn import git <path> <repo>...  # Import git commit history (zero-LLM, offline)
kairn import claude-code <path>    # Import Claude Code session history (zero-LLM, offline)

Importing your history

kairn import git <workspace> <repo>... backfills a Kairn store from one or more local git repositories at $0 - no LLM calls, no network calls. Conventional-commit prefixes map to experience types (fix: -> solution, feat:/refactor:/perf: -> pattern, everything else -> decision); merge commits are skipped. Imported experiences land in a dedicated imported-git namespace, separate from your organic knowledge, so they're always distinguishable and a bad import is fully reversible.

kairn import git ~/brain ~/code/my-project --dry-run   # Preview first
kairn import git ~/brain ~/code/my-project              # Then import for real
kairn import git ~/brain ~/code/proj-a ~/code/proj-b --since 2026-01-01

Idempotent - re-running only imports commits that weren't already imported, so it's safe to run again as a repo's history grows.

Claude Code transcripts

kairn import claude-code <workspace> backfills your Kairn store from your existing Claude Code session history, also at $0 and fully offline. With no --root given it scans ~/.claude/projects (and ~/.claude-secondary/projects if you have a second account); --root PATH is a repeatable override. Imported experiences land in their own imported-claude-code namespace, so they stay distinct from your organic knowledge and a bad import is reversible.

kairn import claude-code ~/brain --dry-run              # Review exactly what would be stored
kairn import claude-code ~/brain                        # Import (prompts once before writing)
kairn import claude-code ~/brain --root ~/other/projects --since 2026-01-01 --yes

What gets stored (coarse mode): one experience per session - the session's title plus your first prompt of that session. This is deliberately a low-detail, high-precision summary rather than a fine-grained per-decision extraction: a zero-LLM rule-based extractor cannot reliably tell a captured decision from ordinary planning chatter, so import claude-code imports a clean session-level pointer instead of noisy fragments. It is not a full transcript archive, and it is not a one-time migration - it is idempotent and meant to be re-run as your history grows.

Privacy. Every stored string is passed through a deterministic secret redactor first (API keys, Authorization/Bearer headers, password=/token=/secret= assignments, common vendor key shapes, private-key blocks, URL-embedded credentials). Tool outputs and tool-call blocks are never read, only your own prompt text. The redactor is defense in depth, not the only control: a real (non-dry-run) run is gated behind an explicit confirmation, and --dry-run shows you the exact post-redaction text before anything is written. Redaction is bounded by its rule set, so --dry-run review before a first real import is recommended; nothing ever leaves your machine.

Configuration

KAIRN_LOG_LEVEL=INFO|DEBUG|WARNING    # Default: WARNING
KAIRN_DB_PATH=~/brain/.kairn         # Default: {workspace}/.kairn
KAIRN_CACHE_SIZE=100                  # LRU cache entries
KAIRN_JWT_SECRET=<your-secret>        # Required for team features

Development

git clone https://github.com/primeline-ai/kairn
cd kairn
pip install -e ".[dev,team]"
pytest tests/ -v --cov
ruff check src/ && ruff format src/

Project Structure

src/kairn/
├── server.py              # FastMCP server + 22 tools
├── cli.py                 # CLI commands
├── config.py              # Configuration
├── core/
│   ├── graph.py           # GraphEngine (6 tools)
│   ├── memory.py          # ProjectMemory (3 tools)
│   ├── experience.py      # ExperienceEngine (4 tools)
│   ├── ideas.py           # IdeaEngine (2 tools)
│   ├── intelligence.py    # IntelligenceLayer (5 tools)
│   └── router.py          # ContextRouter
├── storage/
│   ├── base.py            # Storage interface
│   └── sqlite_store.py    # SQLite + FTS5 implementation
├── models/                # Data models
├── events/                # Event bus
└── auth/                  # JWT + RBAC (team feature)

Performance

Measure it yourself rather than trusting this table:

kairn benchmark ~/brain --nodes 100

One run of that command, 100 nodes, on an Apple M4 Pro:

Single run on one machine, so treat it as a shape rather than a spec - which is why the command is above the table. kn_connect and kn_crossref used to appear here with figures the benchmark does not produce; they have been removed rather than estimated.

Used By

License

MIT

---

Part of the PrimeLine Ecosystem

@PrimeLineAI · primeline.cc · Free Guide