by LING71671
Open-source reverse engineering lab: 197-article knowledge base + MCP tools + CTF/APK/PE automation toolchain. Agent-native. Note:由于场景原因，目前有让几乎所有（除fable5）AI都会越狱的bug,静等官方修复,但是请在合法授权下进行安全测试😉(等我用上5.6再更新(ー_ー)!!)
# Add to your Claude Code skills
git clone https://github.com/LING71671/open-reverselabGuides for using ai agents skills like open-reverselab.
Last scanned: 6/28/2026
{
"issues": [],
"status": "PASSED",
"scannedAt": "2026-06-28T07:51:29.463Z",
"npmAuditRan": true,
"pipAuditRan": true,
"promptInjectionRan": true
}open-reverselab is an open-source ai agents skill for AI coding assistants such as Claude Code, Codex CLI, and ChatGPT, built by LING71671. Open-source reverse engineering lab: 197-article knowledge base + MCP tools + CTF/APK/PE automation toolchain. Agent-native. Note:由于场景原因，目前有让几乎所有（除fable5）AI都会越狱的bug,静等官方修复,但是请在合法授权下进行安全测试😉(等我用上5.6再更新(ー_ー)!!). It has 681 GitHub stars.
Yes. open-reverselab passed SkillsLLM's automated security scan — a dependency vulnerability audit plus prompt-injection heuristics — with no high-severity issues. You can read the full report in the Security Report section on this page.
Clone the repository with "git clone https://github.com/LING71671/open-reverselab" and add it to your Claude Code skills directory (see the Installation section above).
open-reverselab is primarily written in Python. It is open-source under LING71671 on GitHub, so you can review or fork the full source.
Yes. SkillsLLM lists many other AI Agents skills you can browse and compare side by side. Open the AI Agents category from the badge at the top of this page, or use the Related Skills and comparison links further down to weigh open-reverselab against similar tools.
No comments yet. Be the first to share your thoughts!
Open-source reverse engineering lab — 178-article knowledge base, 100+ MCP automation tools, covering CTF pentesting / APK reverse engineering / PE binary analysis / cryptography & protocol cracking / game cheating analysis. Agent-native, directory-as-convention.
Signal → kb_router(board=) → kb_read_file → Attack chain → MCP tool mapping → Execution
| Signal Type | Board | KB Categories / Files | MCP Tool Family |
|---|---|---|---|
| HTTP/Web/API/CVE/Cloud/CAPTCHA | ctf-website |
26/118 | http_probe run_ctf_tool kb_router |
| APK/DEX/SO/Frida/Java | apk-reverse |
8/20 | android_app_baseline android_crypto_unpack_recipe android_frida_* |
| PE/x64/x86/malware/driver | pe-reverse |
9/22 | triage_pe ghidra_headless_analyze make_x64dbg_breakpoint_script sample_full_workup |
| Crypto/Protocol/Cheat/IoT/Radio | general |
5/17 | die_scan ghidra_* rizin_* python_re_tool_* |
kb/
├── ctf-website/techniques/ 26 categories, 118 articles — Full web attack surface
├── apk-reverse/techniques/ 8 categories, 20 articles — APK/DEX reverse engineering
├── pe-reverse/techniques/ 9 categories, 22 articles — PE binary analysis
└── general/techniques/ 5 categories, 17 articles — Cryptography / Protocols / Kernel / Cheating / Methodology
Each technique file follows this structure: Scenario → Input signal → Method → Attack chain → MCP tool mapping
Agent workflow: detect signal → kb_router lookup → kb_read_file → execute via MCP tool mapping.
| Board | Trigger Signals |
|---|---|
boards/ctf-website |
URL, HTTP, JWT, SQLi, SSRF, CVE, API, CSP, OAuth, CAPTCHA, Cloudflare, ReDoS, Slowloris, DoS, Paywall |
boards/android |
APK, DEX, adb, Frida, jadx, smali, SO, native |
boards/windows |
PE, EXE, DLL, x64dbg, Ghidra, Procmon, packer, malware |
boards/general |
AES/DES/RSA, protobuf, game cheat, EAC/BE/Vanguard, firmware, JTAG, SDR |
boards/misc |
MCP config, skill installation, environment health check |
samples/ → Original samples + _quarantine/ + unpacked/
exports/ → Tool outputs (triage / IOC / YARA / Sigma / Procmon / Ghidra summaries)
patches/ → Patch artifacts (original samples are never modified)
notes/ → Analysis notes
reports/ → Final reports
scripts/ → Automation scripts
projects/ → Ghidra project files
templates/ → Note / report / rule templates
kb/ → Reusable attack knowledge base
tools/ → Toolchain
cases/ → Lightweight index — no large file copies
On Windows, beginners can double-click START_HERE.bat or START_HERE.cmd
from the repository root. It checks Python, uv, Git, workspace layout, and
reverse_lab_tools MCP; creates core wrappers; runs real MCP tool calls; gives
install advice for missing items; and writes reports/misc/first-run-report.json
plus reports/misc/mcp-smoke-report.json.
On macOS/Linux, run ./START_HERE.sh from the repository root. It performs the
same first-run checks and uses POSIX shell wrappers under tools/bin/; optional
Windows GUI/PE tools are skipped or reported as Windows-only. Platform-specific
release artifacts can stay separate: Windows full-toolchain releases ship
.bat/PowerShell and GUI tools, while macOS/Linux releases ship the Python,
MCP, shell-wrapper, and native CLI paths.
To have an AI Agent perform setup for you, copy the AI install prompt into Codex or Claude Code. If you are not sure where to start, open START.md.
git clone https://github.com/LING71671/open-reverselab.git
cd open-reverselab
python scripts/misc/first_run_check.py # Check workspace + reverse_lab_tools MCP
uv run --project tools/skills/mcp/ReverseLabToolsMCP python scripts/misc/mcp_smoke_check.py --write-report
.\scripts\misc\bootstrap.ps1 # Core script wrappers (no downloads)
.\scripts\misc\install_tools.ps1 -CTF # Web tools
.\scripts\misc\install_tools.ps1 -Android # APK tools
.\scripts\misc\install_tools.ps1 -Windows # PE tools
.\scripts\misc\install_tools.ps1 -Common # Ghidra + Maven
macOS/Linux quick start:
./START_HERE.sh
./scripts/misc/bootstrap.sh
export PATH="$PWD/tools/bin:$PWD/tools/ctf-website/bin:$PATH"
python scripts/misc/ai_toolcheck.py --board misc
<workspace>/open-reverselab.START_HERE.bat or START_HERE.cmd for the first-run check. macOS/Linux: run ./START_HERE.sh.cd <workspace>/open-reverselab before starting the session.open-reverselab folder directly.python scripts/misc/new_task.py --board ctf-website --name <name>.uv run --project tools/skills/mcp/ReverseLabToolsMCP python scripts/misc/mcp_smoke_check.py --write-report and confirm MCP tool calls pass.Post-install verification:
python scripts/misc/lab_healthcheck.py
python scripts/misc/ai_toolcheck.py --board misc
python scripts/misc/public_release_check.py
--board misc verifies the fresh-clone core Agent scripts and lightweight tools. Run the full python scripts/misc/ai_toolcheck.py only after installing the Android, Windows, and CTF board toolchains you need.
On startup the Agent loads context along this chain:
CLAUDE.md → AGENTS.md → AI-USAGE.md → boards/<board>/AI-USAGE.md
Pair with codex-session-patcher for one-click project-level .codex/ environment and MCP server configuration.
By accessing or using this project, you agree to be bound by the full disclaimer.
The disclaimer covers: all versions and branches (retroactive and prospective), all users (direct and indirect), all derivatives (forks, copies, redistributions), legal compliance across all jurisdictions (including export controls and data protection laws), authorized purposes only, prohibited uses, no warranty, limitation of liability, indemnification, mandatory disclaimer retention in derivatives, anti-removal provisions, and educational communication protections, third-party transaction protections, and unauthorized distribution & impersonation protections, and AI/ML training protections.
📄 Read the full legal disclaimer: DISCLAIMER.md | 中文版
GPL-3.0-only. See LICENSE for details.