Agent-Field/SWE-AF
A runtime that spins up an autonomous AI agent engineering team to scope, build, adapt, and ship software via production-grade PRs.
Not currently ranked — collecting fresh signals.
star history
SWE-AF is an autonomous software engineering factory built on the AgentField framework. It coordinates multiple specialized AI agents — product managers, architects, coders, reviewers, and testers — that work together to take a high-level goal and deliver shipped code. The system supports adaptive control over the factory pipeline and integrates with major LLM providers like Claude Code, Gemini CLI, and OpenRouter to power the agentic workflows.
Frequently asked
- What is Agent-Field/SWE-AF?
- A runtime that spins up an autonomous AI agent engineering team to scope, build, adapt, and ship software via production-grade PRs.
- Is SWE-AF open source?
- Yes — Agent-Field/SWE-AF is open source, released under the Apache-2.0 license.
- What language is SWE-AF written in?
- Agent-Field/SWE-AF is primarily written in Go.
- How popular is SWE-AF?
- Agent-Field/SWE-AF has 941 stars on GitHub.
- Where can I find SWE-AF?
- Agent-Field/SWE-AF is on GitHub at https://github.com/Agent-Field/SWE-AF.