forex-edge
Fail
Audited by Snyk on Mar 27, 2026
Risk Level: HIGH
Full Analysis
HIGH W007: Insecure credential handling detected in skill instructions.
- Insecure credential handling detected (high risk: 1.00). The prompt asks the user for MT5 credentials (server/login/password) and instructs running commands that embed the password as a CLI argument (e.g., --password {PASSWORD}) and saving it in config.json, which forces the agent to handle and output secrets verbatim — high exfiltration risk.
MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).
- Third-party content exposure detected (high risk: 0.80). The SKILL.md explicitly instructs cloning a public GitHub repo (git clone https://github.com/Hero988/Forex-Trading-Skill.git) and then running scripts from that repo (e.g., scripts/*.py like run_full_backtest.py, generate_ea.py), so the agent fetches and executes public third‑party content which can materially influence decisions and tool use.
MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).
- Potentially malicious external URL detected (high risk: 1.00). The skill clones and later runs scripts from the external repository at runtime (git clone https://github.com/Hero988/Forex-Trading-Skill.git .), so fetched remote code would be executed locally and thus directly controls the agent's behavior.
MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).
- Direct money access detected (high risk: 1.00). The skill is explicitly built to produce and deploy MetaTrader 5 Expert Advisors that autonomously open, size, manage, and close live trades. It requests MT5 server/login/password, provides mt5_connect.py and scripts to pull historical data, generate .mq5 EA files, compile/install them, and contains EA features like risk-based position sizing, market entry/exit logic, emergency close, kill switch, and trade management (T1/T2, trailing stops, float cap). The prompt even states the generated EAs "trade on MetaTrader 5 without human intervention." These are direct market-order execution capabilities, so this is direct financial execution authority.
Issues (4)
W007
HIGHInsecure credential handling detected in skill instructions.
W011
MEDIUMThird-party content exposure detected (indirect prompt injection risk).
W012
MEDIUMUnverifiable external dependency detected (runtime URL that controls agent).
W009
MEDIUMDirect money access capability detected (payment gateways, crypto, banking).
Audit Metadata