MindJack

MindJack

Get into the mind of AI agents. Read their memories. Rewrite their instructions.

Security research toolkit for auditing the local attack surface of AI coding assistants.

**Zero dependencies, Python 3.10+ standard library only.**

1. extractor.py - Extract conversation history from 10 AI tools
2. injector.py - Scan injection surfaces, demonstrate 56 attack scenarios
3. mindjack report - Generate interactive HTML attack graph with OWASP compliance mapping

**Disclaimer:** For authorized security testing, red teaming, CTF challenges, and AI safety research only. Always obtain proper authorization before testing on systems you don't own.

---

Supported Tools

Tool	Extractor	Injector	Key Attack Surface
Claude Code	history, sessions, memories	rules, settings, hooks, MCP, memory	CLAUDE.md, settings.json, hooks (RCE)
OpenAI Codex CLI	history, rollout sessions	AGENTS.md, config.toml	sandbox_mode: danger-full-access
GitHub Copilot	chat sessions	copilot-instructions.md	.github/copilot-instructions.md
Cursor IDE	state.vscdb chat data	.cursorrules, MCP config	CVE-2025-54135 (MCP RCE)
Aider	chat history markdown	config.yml, CONVENTIONS.md	auto-commits, model override
Continue.dev	session JSON	config.yaml, rules, .continuerc.json	config.ts (arbitrary code exec)
Cline / Roo Code	task conversations	.clinerules, .roo/rules/, memory-bank	Persistent memory bank poisoning
Amazon Q	chat history JSON	.amazonq/rules/, MCP config	MCP server injection
Windsurf/Codeium	cascade, memories	.windsurfrules, global_rules.md	Auto-generated memory poisoning

---

Quick Start

git clone https://github.com/7h30th3r0n3/MindJack.git
cd MindJack

# Extract all conversations
python3 extractor.py

# Scan for injection surfaces
python3 injector.py scan

# List a