Hector
Hector is an open-source AI agent runtime that runs as a single self-contained binary with a built-in web UI. It solves the problem of deploying and managing AI agents on your own infrastructure without external dependencies or mandatory cloud accounts. It is for teams that need full control, on-premise or air-gapped deployment, and multi-tenancy.
✨ Key features
- Single ~30MB Go binary with embedded Studio UI
- Zero external dependencies, no runtime interpreters
- Config-driven YAML for agents, RAG, guardrails, triggers
- SQL-backed persistence (SQLite or PostgreSQL)
- Enterprise security: OIDC, PII redaction, sandboxing
- Multi-agent orchestration patterns (sequential, parallel, loop)
🎯 Use cases
- Deploy AI agents on-premise or in air-gapped networks
- Build multi-agent workflows with orchestration patterns
- Create RAG pipelines with document ingestion and retrieval
- Automate tasks via scheduled or webhook triggers
- Implement human-in-the-loop approval workflows
📦 Installation
🧰 Requirements: Runs on Linux, macOS, or Windows; no external dependencies; optional LLM API keys for providers like Anthropic, OpenAI, Gemini, or Ollama.
macOS / Linux
curl -fsSL https://gohector.dev/install.sh | sh
hector serve
Windows (PowerShell)
irm https://gohector.dev/install.ps1 | iex
hector serve
Homebrew
brew install verikod/tap/hector
hector serve
Docker
docker run -p 8080:8080 ghcr.io/verikod/hector:latest serve
🚀 Usage
After installation, run hector serve and open http://localhost:8080/. An admin secret is printed in the terminal; enter it to unlock Studio and configure LLM providers, create agents, and start chatting. No config files needed.
❓ FAQ
What are the system requirements to run Hector?
Hector is a single binary that works on Linux, macOS, and Windows. It has no external dependencies or runtime interpreters.
How do I configure LLM providers?
You can configure providers via YAML config or through the Studio UI. Supported providers include Anthropic, OpenAI, Gemini, and Ollama.
Can I deploy Hector in an air-gapped environment?
Yes, Hector is self-sovereign by design and can be deployed on-premise or in air-gapped networks. LLM keys stay on your machines and data never transits third-party infrastructure.
How does Hector handle persistence?
Hector uses SQL-backed persistence with SQLite (default) or PostgreSQL. Sessions, tasks, and checkpoints are stored and survive restarts.
📊 Repository
🤖 Overview, features, install steps and FAQ were generated from the project's README on Sep 4, 2026. Always check the original source before running commands.