# SWE-agent/SWE-agent

SWE-agent takes a GitHub issue and tries to automatically fix it, using your LM of choice. It can also be employed for offensive cybersecurity or competitive coding challenges. [NeurIPS 2024]

Repository: https://github.com/SWE-agent/SWE-agent
Canonical: https://ross.abutalabs.com/products/swe-agent
Homepage: https://swe-agent.com
Language: Python
License: MIT
License Family: permissive
Topics: agent, ai, developer-tools, llm, agent-based-model, lms, cybersecurity
Last push: 2026-08-24T22:03:09+00:00

## Health v2 (maintenance only)
Score: 67/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 28, longevity 63
- inputs: {"age_days": 883, "days_push": 9, "days_rel": 468, "gap_med": 82, "n_releases_24m": 4}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 20147, forks 2206 (observed 2026-08-28T04:11:29.752800+00:00)

## What it is
SWE-agent is a Python framework that lets a language model autonomously use tools to fix GitHub issues, find cybersecurity vulnerabilities, or solve coding challenges. It achieved state-of-the-art results on SWE-bench and is configurable via a single YAML file.

## Use cases
- automatically fix github issues with an llm
- run ai agent on swe-bench benchmark
- find cybersecurity vulnerabilities with an ai agent
- solve competitive coding challenges with an llm agent
- build a custom autonomous coding agent
- research on language model agents

## When to choose
- you want an LLM to autonomously resolve real repository issues
- you need a hackable, research-friendly agent framework with SWE-bench SOTA results
- you want to benchmark or evaluate LLMs on software engineering tasks

## When to avoid
- you want the team's actively developed successor - use mini-swe-agent instead
- you need a production CI bot rather than a research/benchmark tool
- you want a fully managed commercial service

## Facets
- artifact type: framework
- maturity: maintenance
- function: agent-framework, llm-inference, developer-tools
- domain: large-language-models, developer-tools, security
- platform: python, cli
- tags: swe-bench, autonomous-agents, automated-bug-fixing, cybersecurity, research, ai-agents, docker

## Member repositories
- SWE-agent/SWE-agent (main) score 67

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:11:29.752800+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-29T16:59:20.027772+00:00, confidence not recorded.
  - readme: https://github.com/SWE-agent/SWE-agent (fetched 2026-08-28T04:11:29.752800+00:00, sha fb021a1fb871)
  - homepage: https://swe-agent.com (fetched 2026-08-29T07:57:43.429005+00:00, sha 36c6c3c2e4f9)
- Data as of 2026-08-30T08:39:29.467469+00:00.