Vela · a self-hosted agent harness logo

Vela · a self-hosted agent harness

Free

Vela is a self-hosted agent harness in one static Go binary — swappable model, Discord front end, optional build and deploy plugins.

Visit Website

At a glance

Explore the listed capabilities, then check the details that matter to you.

  • Self-hosted AI agent harness
  • Single static Go binary
  • Swappable AI model
ai-agentdeepseekdiscord-botembeddedgolangaiai assistantautomation
Vela · a self-hosted agent harness screenshot 1

Overview

Vela is a self-hosted agent harness distributed as one static Go binary. The binary holds the agent loop, the tool belt, and persistent memory. It can be pointed at any OpenAI-compatible model and given a chat front end, and it runs on a server, a laptop, or a credit-card-sized RISC-V board drawing two watts.

Installation centers on the harness: the agent loop, tools, and memory in one binary, with a swappable model and Discord as the shipped front end. Optional plugins include Holodex, described as a deploy target, and the Foundation, a Rails 8.1 template for generated apps with Material Design 3, auth, commerce, and validated boot-time configuration.

The agent loop provides tool-calling with phase boundaries and prompt-injection guards. Tools include search, read, charts, images, GitHub, and shell, gated per user. Memory includes per-channel history and long-term notes stored as plain files on disk. Plugins can attach a build target and a template when apps are wanted. Vela is MIT licensed, self-hosted, uses your keys and your box, and includes AGENTS.md for coding agents.

Pricing

Free

Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.

Visit Website

Prices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.

Use Cases

  • Teams use Vela to run an agent on a server, laptop, or low-power RISC-V board.
  • Teams use Vela to connect an OpenAI-compatible model to a chat front end.
  • Teams use Vela to deploy with Discord as the shipped chat front end.
  • Teams use Vela to run tool-calling with phase boundaries and prompt-injection guards.
  • Teams use Vela to give users gated access to search, read, charts, images, GitHub, and shell tools.
  • Teams use Vela to keep per-channel history and long-term notes as plain files on disk.
  • Teams use Vela to attach a build target and template when they want generated apps.
  • Teams use Vela to provide AGENTS.md for coding agents.

Features

Self-hosted AI agent harness

Single static Go binary

Swappable AI model

Discord front end

Optional build plugins

Optional deploy plugins

Reviews

No reviews yet. Be the first to share your experience!

Rankings & collections