open-reverselab - Open-source AI reverse-engineering agent platform

项目说明

<di="center">

Open-Re — AI Reverse Engineering Agent & MCP Lab

Open-source AI-powered re platform for autonomous binary analysis, built for Claude Code, Codex, OpenCode and any MCP-compatible agent.

Ghidra · Frida · x64dbg · Rizin · PE · APK · Malware Analysis · CTF · Binary Analysis

---

An open-source re lab — executable knowledge base, 100+ MCP tools, Agent-native.

*From an input signal to an e chain, every step is runnable.*

<br />

![Discord]()
![Ask DeepWiki]()

English · 简体中文

</div>

---

:handshake: Sponsored by Sentry

<di="center">

Sentry supports openre with a sponsored account — error monitoring and performance tracing for the lab's toolchain.

<picture>
<source media="(prefers-color-scheme: dark)" srcset="assets/sponsors/sentry-wordmark-dark.svg">
<img src="assets/sponsors/sentry-wordmark-light.svg" width="220" alt="Sentry">
</picture>

_{Since Aug 2026 — see <a href="SPONSORS.md">SPONSORS.md</a> for the full sponsor list and how to become one.}

</div>

---

What is Re

Re is an opinionated, runnable attack-knowledge base for reverse engineers, security researchers, CTF players, and AI Agents. Every article is structured as Scenario → Input signal → Method → Attack chain → MCP tool mapping, so a human or an Agent can pick up at any entry signal and walk the chain to e.

The lab is built to be run, not browsed. If a step can't be executed, it doesn't belong in the KB.

Who is this for

| You are … | Start here |
| --- | --- |
| A CTF player stuck on a web/Android/PE challenge | Browse the CTF Website, APK Reverse, or PE Reverse board |
| A security researcher triaging a sample | Quick start → boards/<board>/AI-USAGE.md → sample_full_workup MCP tool |
| A reverse engineer who wants a workspace, not a tutorial | Repository layout — clone and start |
| An AI Agent de wiring up re tools | For AI Agents — context chain, MCP smoke test, en protocol |

Quick links

<di="center">

:book: Knowledge Base
&nbsp;&nbsp; :rocket: Quick start
&nbsp;&nbsp; :handshake: Contributing
&nbsp;&nbsp; :speech_balloon: Discord

</div>

---

Knowledge at a glance

| Board | Trigger signals | MCP entry points |
| --- | --- | --- |
| ctf-website | URL · HTTP · JWT · SQLi · SSRF · CVE · API · CSP · OAuth · CAPTCHA · Cloudflare · ReDoS · Slowloris · DoS · Paywall | http_probe, run_ctf_tool, kb_router |
| apk-reverse | APK · DEX · adb · Frida · jadx · smali · SO · native | android_app_baseline, android_crypto_unpack_recipe, android_frida_* |
| pe-reverse | PE · EXE · DLL · x64dbg · Ghidra · Procmon · packer · malware | triage_pe, ghidra_headless_analyze, make_x64dbg_breakpoint_script, sample_full_workup |
| general | AES · DES · RSA · protobuf · game cheat · EAC / BE / · firmware · JTAG · SDR | die_scan, ghidra_*, rizin_*, python_re_tool_* |
| misc | MCP config · skill install · en check | mcp_smoke_check, ai_toolcheck, lab_healthcheck |

The complete directory tree is a under kb/.

---

Quick start

Pick the path that matches your role, not your platform — the platform differences are below the role split.

For humans

Windows (recommended for first run) — double-click START_HERE.bat (or START_HERE.cmd) in the repo root. It checks Python, uv, Git, the re MCP server, runs real MCP smoke calls, and writes reports/misc/first-run-report.json plus reports/misc/mcp-smoke-report.json.

macOS / Linux — run ./START_HERE.sh from the repo root. It does the same first-run checks using POSIX shell wrappers under tools/bin/. Windows-only GUI/PE tools are skipped or reported as such.

<details>
<summary><b>Per-board install (after first-run passes)</b></summary>

macOS / Linux equi:

Install only the boards you need. Don't dump the full toolchain if you're only doing web CTF.

</details>

<details>
<summary><b>Windows Defender / antivirus note</b></summary>

After installing CTF / ExploitDB tools, Windows Defender may flag exploit samples or payload documentation — e.g. tools/ctf-website/exploitdb, kb/ctf-website/techniques/24-database/03-nosql-injection.md, docs/llms-full.txt. These contain security test payloads, webshells, shellcode, or ExploitDB samples and are expected components.

We recommend minimal path exclusions rather than excluding the entire repo:

If indi documents are flagged, handle them case-by-case.

</details>

For AI Agents

1. Clone into a dedicated local directory, e.g. <workspace>/open-re.
2. Claude Code: cd <workspace>/open-re, then start your session.
3. Codex APP: Open the existing open-re folder directly (no re-clone needed).
4. Automated AI install: send the contents of templates/prompts/ai-install.en.md to your Agent.
5. Create a task: python scripts/misc/new_task.py --board ctf-website --name <name>.
6. real MCP in whene machines or updating configs:

Context chain — Agents load context through this sequence:

Use codex-session-patcher for one-click configuration of project-level .codex/ and MCP servers.

<details>
<summary><b>En snapshot protocol (host-level, shared across projects)</b></summary>

When first opening this project with an AI Agent, it inspects your local en (OS, toolchains, re libraries, devices, sanitized env vars, network, workspace) following the protocol in AGENTS.md, and caches a snapshot at ~/.open-re/env/env.md (%USERPROFILE%\.open-re\env\env.md on Windows).

This file is host-level and shared across sessions: subsequent sessions read it directly without re-probing, unless 7 days ha or the protocol has bumped. It remains local and is ne. Sensiti are redacted (keys marked "SET", proxies stripped of credentials).

</details>

Post-install

--board misc checks the fresh-clone core Agent scripts and lightweight tools. Run full python scripts/misc/ai_toolcheck.py only after installing Android / Windows / CTF boards.

---

Repository layout

The repository uses directory-as-con. Put artifacts where they belong and the tooling finds them automatically.

> Don't commit large samples, full PCAPs, or memory dumps. Index them in cases/ and link to a private store.

---

Community & contributing

---

Sponsors

Re is sustained by people and organizations that fund the infrastructure, tooling, and review time.

<table width="100%">
<thead>
<tr>
<th width="180" align="left">Logo</th>
<th align="left">Sponsor / 赞助方</th>
<th width="110" align="left">Since / 时间</th>
<th align="left">Support / 支持内容</th>
</tr>
</thead>
<tbody>
<tr>
<td align="left" ="middle">
<a href="
<source media="(prefers-color-scheme: dark)" srcset="assets/sponsors/sentry-wordmark-dark.svg">
<img src="assets/sponsors/sentry-wordmark-light.svg" width="150" alt="Sentry">
</picture></a>
</td>
<td ="middle"><strong><a href="
<td ="middle">2026-08</td>
<td ="middle">Sponsored account — error monitoring &amp; performance tracing platform（错误监控与性能追踪平台）</td>
</tr>
</tbody>
</table>

Want to sponsor? See SPONSORS.md or email — we list every sponsor, big or small, with the support they provide.

---

Disclaimer

By accessing or using this project, you agree to be bound by the full disclaimer — co all and branches, all users (direct and indirect), all deri, legal compliance across all jurisdictions (including export controls and data protection laws), authorized purposes only, prohibited uses, no warranty, limitation of liability, mandatory disclaimer retention in deri, AI/ML training protections, and more.

> :page_facing_up: Full legal text: DISCLAIMER.md · 中文版

License

GPL-3.0-only. See LICENSE. By contributing, you agree your contributions are licensed under the same terms.

项目信息

开发语言：Python

Star 数：1214

技术要点

该项目属于病毒分析方向的开源资源，代码结构清晰，适合安全研究人员学习参考。项目围绕Open-source AI reverse-engineering agent platform and MCP server for Ghidra, Frida, x64dbg and Rizin — automated PE/APK/binary analysis, CTF and malware research, with 100+ MCP tools and a 194-article runnable knowledge base.展开，提供了可运行的实现，对理解相关安全机制、工具原理与攻防技术具有较高的研究价值。

从技术实现角度看，此类项目的核心在于对底层原理的深入理解与工程化落地。开发者在研究过程中可以结合官方文档与社区讨论，梳理项目的初始化流程、核心模块职责与数据流转，逐步掌握其中的技术细节与设计思路。

使用与学习建议

1. 下载源码压缩包，解压到本地环境。

2. 根据项目类型准备对应的运行环境与依赖。

3. 按照 README 中的步骤完成配置与部署。

4. 在隔离的测试环境中运行验证，避免影响生产系统。

免责声明

本资源来源于开源社区，仅供安全研究、技术学习与合法授权范围内的攻防演练使用，请勿用于任何非法或侵害他人权益的用途。使用本项目产生的一切后果由使用者自行承担。