Nestor Documentation

Nestor is an open-source AI agent platform designed for developers who need autonomous agents they can actually trust. It provides a complete stack from CLI to web dashboard, with security built into every layer.

Key Capabilities

Mission System — Give an objective, get a structured report. Sub-agents decompose, execute, cross-reference, iterate
Self-Improving Agents — Agents create their own tools on-the-fly via nestor_build_tool
Knowledge Graph — Second brain that grows with every mission. Entities, relationships, facts
7 LLM Providers — Claude, GPT-4o, Gemini, Grok, Mistral, Ollama, and OpenRouter (300+ models)
30+ Agent Tools — Web search, scraping, OSINT, file ops, shell, Obsidian, n8n, memory, missions
Hat System — 5 specialized personas (coder, researcher, security, writer, osint) with role-specific tools
Backpressure Validators — Tests as quality gates with progressive strictness for loop commands
Desktop Notifications — OS-native alerts when missions complete or agents need attention
Rust Security Core — SSRF protection, path traversal prevention, secret redaction via N-API
Docker Sandbox — Isolated execution with cap-drop ALL and read-only filesystem
DAG Workflows — Multi-agent orchestration with parallel execution and governance gates
Trust Scores — A-F grading based on real execution history
Obsidian & n8n Integration — Connect your knowledge base and automation workflows
Studio Dashboard — 18+ page web UI for monitoring, editing workflows, and tracking costs

Architecture

Nestor is built as a monorepo with distinct packages, each with a clear responsibility:

Package	Role
`@nestor/cli`	Command-line interface with 12+ commands
`@nestor/agent`	Agent runtime, LLM adapters, tool execution, missions
`@nestor/server`	Express + WebSocket server for Studio (18+ pages)
`@nestor/orchestrator`	DAG workflow engine with cron scheduling
`@nestor/mcp`	Model Context Protocol — 30+ tools and 4 resources
`@nestor/db`	SQLite database with 14 tables
`@nestor/skill-tester`	Skill testing framework with mock LLM
`nestor-core`	Rust security module compiled to N-API

CLI Quick Reference

# Installation
npx nestor-sh install

# Agent management
npx nestor-sh agent create --name coder --adapter claude
npx nestor-sh agent list
npx nestor-sh agent config coder

# Interactive shell
npx nestor-sh shell
npx nestor-sh shell --agent coder --model claude-sonnet-4-6
npx nestor-sh shell --hat osint

# Web Studio
npx nestor-sh start

# Missions
npx nestor-sh mission list
npx nestor-sh mission view <id>

# Workflows
npx nestor-sh workflow run code-review.yaml
npx nestor-sh workflow list

# Testing
npx nestor-sh test
npx nestor-sh test --file tests/my-skill.test.yaml
npx nestor-sh test --verbose

Next Steps

Follow the Quickstart Guide to set up your first agent
Learn about agent configuration, 7 LLM providers, and the Hat System
Understand the Mission system for structured investigations
Explore DAG workflows for multi-agent orchestration
Review the security model, circuit breaker, and trust scoring
Set up integrations with Obsidian, n8n, and OpenRouter

✎ Edit this page on GitHub · Last updated 2026-04-26