S I G N E T

Portable Agent Infrastructure

Website · Docs · Vision · Discussions · Discord · Contributing · AI Policy

Your agent is an investment. Signet is where its value accumulates.

Signet is a persistent cognition layer for AI agents. It gives your agent memory that works the way memory actually works — ambient, automatic, and not dependent on the agent deciding to remember. Your agent doesn't call a "save memory" tool. It doesn't search a database when it needs context. Signet extracts knowledge after sessions, builds a knowledge graph, and injects the right context before every prompt. The agent just has its memory. Like you have yours.

Everything runs locally. You own the data. The agent is yours.

bun add -g signetai        # or: npm install -g signetai
signet setup               # interactive setup wizard

The wizard initializes ~/.agents/, configures your harnesses, sets up an embedding provider, creates the database, and starts the daemon.

Paste this to your AI agent:

Install and fully configure Signet AI by following this guide exactly: https://signetai.sh/skill.md

signet status              # check daemon health
signet dashboard           # open the web UI

signet remember "prefers bun over npm"
signet recall "coding preferences"

In connected harnesses, skills work directly:

/remember critical: never commit secrets to git
/recall release process

At the end of every conversation, a distillation engine reviews the session and extracts structured insights — no tool calls, no agent involvement. A knowledge graph maps how those insights connect. A predictive scorer, trained on your interaction patterns, injects the right context before every prompt.

Your agent doesn't manage its memory. It just has it.

Read more: Why Signet · Architecture · Knowledge Graph · Pipeline

CLI (signet)
  setup, knowledge, secrets, skills, hooks, git sync, service mgmt

Daemon (@signet/daemon, localhost:3850)
  |-- HTTP API (90+ endpoints across 18 domains)
  |-- Distillation Layer
  |     extraction -> decision -> graph -> retention
  |-- Predictive Scorer
  |     entity-weight traversal, per-user trained model
  |-- Document Worker
  |     ingest -> chunk -> embed -> index
  |-- Auth Middleware
  |     local / team / hybrid, RBAC, rate limiting
  |-- File Watcher
        identity sync, git auto-commit

Core (@signet/core)
  types, identity, SQLite, hybrid search, graph traversal

SDK (@signet/sdk)
  typed client, React hooks, Vercel AI SDK middleware

Connectors
  claude-code, opencode, openclaw

• Quickstart
• CLI Reference
• Configuration
• Hooks
• Harnesses
• Secrets
• Skills
• Auth
• Dashboard
• SDK
• API Reference
• Knowledge Architecture
• Knowledge Graph
• Spec Index

git clone https://github.com/Signet-AI/signetai.git
cd signetai

bun install
bun run build
bun test
bun run lint

cd packages/daemon && bun run dev        # Daemon dev (watch mode)
cd packages/cli/dashboard && bun run dev # Dashboard dev

Requirements: Node.js 18+, Bun, Ollama (recommended) or OpenAI API key. macOS or Linux.

See CONTRIBUTING.md. Build on existing patterns. Open an issue before contributing significant features. Read the AI Policy before submitting AI-assisted work.

Apache-2.0.

signetai.sh · docs · spec · discussions · issues