Vibeleaderboard
← Back to Tools
Index / agent

SWE-agent

github.com/swe-agent/swe-agent
Visit swe-agent.com
Category
AI Agents
Rank
No. 749Tools index
Pricing
Open Source
Platform
cli · web
Type
AGENT
Builder
swe-agent
GitHub
20.2k stars
Latest release
v1.1.0
Date

About

SWE-agent lets a language model like GPT-4o or Claude autonomously use tools to fix real GitHub issues, solve custom coding tasks, or find cybersecurity vulnerabilities via its EnIGMA mode. It is a research project from Princeton and Stanford that achieves state-of-the-art results on SWE-bench and is fully configurable through a single YAML file.

What it can do

  • Autonomously resolve GitHub issues using an LLM-driven agent

    GitHub issue → Code fix/patch

  • Find cybersecurity vulnerabilities via EnIGMA mode

    CTF challenge or codebase → Identified vulnerability

  • Configure agent behavior through a single YAML file

    YAML configuration file → Configured agent

Why it made the leaderboard

It lets you point any LLM at a real GitHub issue or CTF challenge and have it edit, run, and test code autonomously, all configured through a single YAML file. Its agent-computer interface design — purpose-built file/search commands instead of raw shell — is the pattern most modern coding agents inherited.

Tags

swe-agentswe-benchcoding-agentautonomous-agentllmcybersecurityctfresearch

Tech Stack

Python

Comments (0)

No comments yet

Editorially curated, with community endorsements as a secondary signal. Corrections welcome.