SWE-agent
swe-agent.com
Open-source autonomous agent for resolving GitHub issues.
Quick Information
- Category:
- Coding
- Pricing:
- Free
- Platforms:
- CLI, GitHub
- Free Plan:
- Yes
- API Available:
- Yes
- Last Verified:
- September 2026
What Is SWE-agent?
SWE-agent is an open-source autonomous agent for resolving GitHub issues, enabling language models like GPT-4o or Claude Sonnet 4 to autonomously use tools to fix issues in real GitHub repositories, find cybersecurity vulnerabilities, or perform custom tasks — the main repository remains actively maintained with development activity as recent as September 22, 2026. Worth understanding: the same team's development effort has increasingly focused on mini-swe-agent, a simpler, 100-line implementation that matches SWE-agent's performance (scoring over 74% on SWE-bench verified) with radically less complexity — the general recommendation from the project itself is now to use mini-SWE-agent going forward for new projects.
This sibling-project relationship is genuinely worth understanding clearly: SWE-agent itself hasn't been abandoned or superseded by an unrelated competitor — rather, the same team built a simpler, equally effective successor within the same project family, and the original SWE-agent remains actively maintained (with recent commits and open issue tracking) for existing users or those wanting its more configurable original implementation.
It's aimed at developers and researchers wanting an open-source autonomous coding agent for GitHub issue resolution or security research, with the project's own recommendation to consider mini-swe-agent for new projects worth taking seriously given its simpler implementation and matched performance — SWE-agent itself remains a solid, actively-maintained choice for users already invested in its more configurable approach or specific existing workflows. See our Coding & Development tools directory for related tools like Sweep AI and OpenAI Codex.
What Can You Use SWE-agent For?
Autonomous GitHub issue resolution
Autonomously fixes issues in real GitHub repositories using an LLM of choice.
Cybersecurity vulnerability research
Finds security vulnerabilities as an offensive cybersecurity research tool.
Competitive coding challenges
Solves competitive coding challenges autonomously.
SWE-agent Features
- Model-agnostic agent framework — Works with GPT-4o, Claude Sonnet 4, or other LLMs of choice.
- Autonomous tool use — Uses tools autonomously to fix issues in real repositories.
- mini-swe-agent (sibling project) — A simpler 100-line implementation matching SWE-agent’s performance, recommended for new projects.
Who Is SWE-agent For?
- Developers — Uses an open-source autonomous agent to resolve GitHub issues without manual coding.
- Security researchers — Uses the framework for offensive cybersecurity vulnerability research.
SWE-agent Pros & Limitations
What We Like
- Genuinely actively maintained with recent development activity as of September 2026
- Model-agnostic design works with multiple leading LLMs of choice
- Sibling project mini-swe-agent offers a simpler, equally performant alternative from the same team
Limitations
- Project itself now recommends mini-swe-agent over SWE-agent for new projects specifically
- Requires bringing your own LLM API access, adding a separate cost beyond the tool itself
- Open-source, community-maintained nature means no official commercial support
SWE-agent Pricing
| Plan | Price | Main Features |
|---|---|---|
| Open Source | Free | Self-hosted, bring-your-own LLM API access |
Pricing was last verified September 2026. Check the official website for the latest pricing.
AI Listing Jungle Ranking
- Global Rank:
- #372
- Overall Score:
- 58.3/100
- Monthly Users:
- –
- Rank Change:
- ▼ 27
This ranking reflects AI Listing Jungle's own independent scoring methodology and is not an endorsement by SWE-agent or its company. How Our Rankings Work
Frequently Asked Questions
Is SWE-agent still maintained?
Yes, with development activity as recent as September 22, 2026.
What is mini-swe-agent?
A simpler, 100-line sibling project from the same team, now recommended for new projects.
Is SWE-agent free?
Yes, it’s fully open-source, though you provide your own LLM API access.
What can SWE-agent do?
Autonomously fix GitHub issues, find security vulnerabilities, and solve competitive coding challenges.
Our Take
SWE-agent remains genuinely actively maintained, but its own project team now recommends the simpler, equally performant mini-swe-agent for new projects — worth considering that sibling tool if starting fresh.
Both remain open-source and free, requiring only your own LLM API access as the actual usage cost.
SWE-agent Alternatives
SWE-agent Comparisons
Ready to try SWE-agent?
Visit the official website to get started.
Visit Official WebsiteAI Listing Jungle is an independent AI tools directory. Product names, logos, and trademarks belong to their respective owners.