agent-knowledge

Cross-session memory and recall for AI coding assistants -- works with Claude Code, Cursor, OpenCode, Cline, Continue.dev, and Aider out of the box. Git-synced knowledge base, hybrid semantic+TF-IDF search, auto-distillation with secrets scrubbing.

Knowledge base with category filtering	TF-IDF ranked session search

Why

AI coding sessions are ephemeral. When a session ends, everything it learned -- architecture decisions, debugging insights, project context -- is gone. The next session starts from scratch.

agent-knowledge solves this with two complementary systems:

1. Knowledge Base -- a git-synced markdown vault of structured entries (decisions, workflows, project context) that persists across sessions and machines.
2. Session Search -- TF-IDF ranked full-text search across session transcripts from all your coding tools, so agents can recall what happened before -- regardless of which tool was used.

Supported Tools

Sessions from all major AI coding assistants are auto-discovered -- if a tool is installed, its sessions appear automatically.

No configuration needed. Additional session roots can be added via the EXTRA_SESSION_ROOTS env var (comma-separated paths).

Features

• Multi-tool session search -- unified search across Claude Code, Cursor, OpenCode, Cline, Continue.dev, and Aider sessions
• Hybrid search -- semantic vector similarity blended with TF-IDF keyword ranking
• Git-synced knowledge base -- markdown vault with YAML frontmatter, auto commit and push on writes
• Auto-distillation -- session insights automatically extracted and pushed to git with secrets scrubbing
• Pluggable adapter system -- add support for new tools by implementing the SessionAdapter interface
• Embeddings -- local (Hugging Face), OpenAI, Claude/Voyage, or Gemini providers
• Fuzzy matching -- typo-tolerant search using Levenshtein distance
• 6 search scopes -- errors, plans, configs, tools, files, decisions
• 6 MCP tools -- consolidated action-based interface (knowledge, knowledge_search, knowledge_session, knowledge_graph, knowledge_analyze, knowledge_admin)
• Configurable git URL -- knowledge_admin(action: "config") for runtime setup, persisted at XDG/AppData location
• Cross-machine persistence -- knowledge syncs via git, sessions read from local storage of each tool
• Real-time dashboard -- browse, search, and manage at localhost:3423
• Secrets scrubbing -- API keys, tokens, passwords, private keys automatically redacted before git push
• Knowledge graph -- relationship edges between entries (related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on) with BFS traversal
• Confidence/decay scoring -- entries scored by access frequency and recency; auto-promotion from candidate to established to proven
• Memory consolidation -- TF-IDF duplicate detection on write (warns of similar entries) plus knowledge_analyze(action: "consolidate") for batch dedup scanning
• Reflection cycle -- knowledge_analyze(action: "reflect") surfaces unconnected entries and generates structured prompts for the agent to identify new graph connections
• Auto-linking on write -- new entries automatically linked to top-3 similar existing entries when cosine similarity > 0.7

Quick Start

Install from npm

npm install -g agent-knowledge

Or clone from source

git clone https://github.com/keshrath/agent-knowledge.git
cd agent-knowledge
npm install && npm run build

Windows note: better-sqlite3 requires native compilation. If npm install fails with gyp ERR!, install the C++ build tools:

winget install Microsoft.VisualStudio.2022.BuildTools --override "--add Microsoft.VisualStudio.Workload.VCTools --includeRecommended --passive"

Then re-run npm install. This is only needed when prebuilt binaries aren't available for your Node.js version (e.g. Node 24+).

Option 1: MCP server (for AI agents)

Add to your MCP client config (Claude Code, Cline, etc.):

{
  "mcpServers": {
    "agent-knowledge": {
      "command": "npx",
      "args": ["agent-knowledge"]
    }
  }
}

The dashboard auto-starts at http://localhost:3423 on the first MCP connection.

See Setup Guide for client-specific instructions (Claude Code, Cursor, Windsurf, OpenCode).

Option 2: Standalone server (for REST/WebSocket clients)

node dist/server.js --port 3423

MCP Tools (6)

Knowledge Base

Search

Scopes: errors, plans, configs, tools, files, decisions, all.

Sessions

Knowledge Graph

Relationship types: related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on.

Analysis

Admin

REST API

Architecture

graph LR
    subgraph Storage
        KB[(Knowledge Base<br/>~/agent-knowledge<br/>Git Repository)]
    end

    subgraph Session Sources
        CC[(Claude Code<br/>JSONL)]
        CU[(Cursor<br/>JSONL)]
        OC[(OpenCode<br/>SQLite)]
        CL[(Cline<br/>JSON)]
        CD[(Continue.dev<br/>JSON)]
        AI[(Aider<br/>MD / JSONL)]
    end

    subgraph agent-knowledge
        KM[Knowledge Module<br/>store / search / git]
        AD[Session Adapters<br/>auto-discovery]
        SE[Search Engine<br/>TF-IDF + Fuzzy]
        DS[Dashboard<br/>:3423]
        MCP[MCP Server<br/>stdio]
    end

    subgraph Clients
        AG[Agent Sessions]
        WB[Web Browser]
    end

    KB <-->|git pull/push| KM
    CC --> AD
    CU --> AD
    OC --> AD
    CL --> AD
    CD --> AD
    AI --> AD
    AD --> SE
    KM --> MCP
    SE --> MCP
    KM --> DS
    SE --> DS
    MCP --> AG
    DS --> WB

Knowledge Graph

Entries can be connected via typed, weighted edges stored in a dedicated edges SQLite table. Eight relationship types are supported: related_to, supersedes, depends_on, contradicts, specializes, part_of, alternative_to, builds_on.

• knowledge_graph(action: "link") creates or updates an edge (with optional strength 0-1)
• knowledge_graph(action: "unlink") removes edges (optionally filtered by type)
• knowledge_graph(action: "list") lists edges for an entry or relationship type
• knowledge_graph(action: "traverse") performs BFS traversal from a starting entry to a configurable depth

Auto-linking

When knowledge_write creates or updates an entry, it automatically finds the top-3 most similar existing entries via cosine similarity and creates related_to edges for any pair scoring above 0.7.

Confidence & Decay Scoring

Each knowledge entry has a confidence score tracked in the entry_scores SQLite table. Search results are ranked using:

finalScore = baseRelevance * 0.5^(daysSinceLastAccess / 90) * maturityMultiplier

Entries mature automatically based on access count:

Frequently accessed entries rise in search rankings; stale entries decay over time.

Search Capabilities

TF-IDF Ranking -- results scored by term frequency-inverse document frequency. Rare terms boost relevance. Global index cached for 60 seconds.

Fuzzy Matching -- Levenshtein edit distance with sliding window. Configurable threshold (default 0.7).

Scoped Recall via knowledge_recall:

Testing

npm test              # Run all 352 tests
npm run test:watch    # Watch mode
npm run lint          # Type-check (tsc --noEmit)

Environment Variables

Documentation

• Setup Guide — installation, client setup (Claude Code, OpenCode, Cursor, Windsurf), hooks
• Architecture — source structure, design principles, database schema
• Dashboard — web UI views and features
• Changelog

License

MIT