# SWE-agent — Automated Bug Fixing with AI Agents

> AI agent system from Princeton NLP that autonomously fixes GitHub issues. Navigates codebases, writes patches, and submits pull requests. Resolves 12% of real-world SWE-bench tasks. 15,000+ stars.

## Install

Save as a script file and run:

## Quick Use

```bash
pip install swe-agent
```

```bash
# Fix a GitHub issue automatically
swe-agent run \
  --model claude-sonnet-4-20250514 \
  --issue https://github.com/user/repo/issues/123
```

SWE-agent clones the repo, analyzes the issue, navigates the codebase, writes a fix, and creates a patch.

---

## Intro

SWE-agent is an AI agent system from Princeton NLP that autonomously resolves real-world GitHub issues with 15,000+ GitHub stars. Give it an issue URL and it clones the repository, navigates the codebase, identifies the bug, writes a patch, and validates the fix — achieving 12.5% on the SWE-bench benchmark of real-world software engineering tasks. It uses a custom Agent-Computer Interface (ACI) designed specifically for code navigation and editing. Best for researchers studying AI coding capabilities and teams experimenting with automated bug fixing. Works with: Claude, GPT-4. Setup time: under 5 minutes.

---

## How It Works

### 1. Issue Analysis
SWE-agent reads the GitHub issue and understands:
- What is broken
- Expected vs actual behavior
- Relevant error messages and stack traces

### 2. Codebase Navigation
Using its Agent-Computer Interface (ACI):

```bash
# The agent navigates like a developer
find_file "auth_handler.py"
open_file "src/auth/auth_handler.py"
search_file "validate_token"
goto_line 145
```

### 3. Bug Identification
The agent reads relevant code, traces the logic, and pinpoints the issue:

```
Analyzing: src/auth/auth_handler.py:145
Issue: Token expiry check uses UTC timestamp but comparison uses local time
Fix: Convert both to UTC before comparison
```

### 4. Patch Generation
Writes and applies the fix:

```python
# Before
if token.expiry < datetime.now():

# After (SWE-agent fix)
if token.expiry < datetime.now(timezone.utc):
```

### 5. Validation
Runs existing tests to verify the fix:

```bash
Running test suite...
test_auth_valid_token: PASS
test_auth_expired_token: PASS (was FAIL)
test_auth_refresh: PASS
All tests passing.
```

### Agent-Computer Interface (ACI)
Custom commands designed for code navigation:

| Command | Action |
|---------|--------|
| `find_file` | Search for files by name |
| `open_file` | Open and display file content |
| `search_file` | Search within current file |
| `search_dir` | Search across directory |
| `goto_line` | Jump to specific line |
| `edit` | Edit lines in current file |
| `submit` | Submit the patch |

### SWE-bench Results

| Model | Resolve Rate |
|-------|-------------|
| Claude Sonnet | 12.5% |
| GPT-4 | 10.3% |
| GPT-3.5 | 3.8% |

### Key Stats
- 15,000+ GitHub stars
- Princeton NLP research
- 12.5% SWE-bench resolve rate
- Custom Agent-Computer Interface
- Supports Claude and GPT-4

### FAQ

**Q: What is SWE-agent?**
A: SWE-agent is an AI system from Princeton NLP that autonomously fixes real-world GitHub issues by navigating codebases, writing patches, and validating fixes.

**Q: Is SWE-agent free?**
A: Yes, open-source under MIT license. You need LLM API keys.

**Q: Can SWE-agent fix any bug?**
A: It resolves about 12.5% of real-world software engineering tasks. It works best on well-specified bugs with clear reproduction steps.

---

## Source & Thanks

> Created by [Princeton NLP](https://github.com/princeton-nlp). Licensed under MIT.
>
> [SWE-agent](https://github.com/princeton-nlp/SWE-agent) — ⭐ 15,000+

Thanks to Princeton NLP for advancing the frontier of automated software engineering.

---

<!-- ZH -->

## 快速使用

```bash
pip install swe-agent
swe-agent run --model claude-sonnet-4-20250514 --issue https://github.com/user/repo/issues/123
```

---

## 简介

SWE-agent 是 Princeton NLP 开发的 AI Agent 系统，GitHub 15,000+ stars。自动克隆仓库、分析 issue、导航代码、编写补丁并验证修复。在 SWE-bench 真实软件工程任务基准上达到 12.5% 的解决率。适合研究 AI 编码能力和尝试自动化 bug 修复的团队。

---

## 来源与感谢

> Created by [Princeton NLP](https://github.com/princeton-nlp). Licensed under MIT.
>
> [SWE-agent](https://github.com/princeton-nlp/SWE-agent) — ⭐ 15,000+

---
Source: https://tokrepo.com/en/workflows/ec5f0e4b-ca6b-4379-8ec2-0e0ef63b9fa5
Author: Agent Toolkit