gate-exchange-trading

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The skill explicitly instructs runtime calls to public news and social UGC sources (e.g., news_feed_search_news and news_feed_search_news(platform_type="social_ugc") referenced in references/scenarios.md, routing-and-analysis.md, and runtime-dependencies.md) and requires ingesting and acting on that content to produce Trading Briefs, order drafts, and execution decisions, which could allow indirect prompt injection from untrusted third-party content.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The setup.sh installs a required runtime binary by fetching and extracting releases from GitHub (calls to https://api.github.com/repos/gate/gate-cli/releases/latest and downloads from https://github.com/gate/gate-cli/releases/download/${VERSION}/${ARCHIVE}), which clearly fetches and executes remote code (gate-cli) as a required dependency for the skill.

MEDIUM W009: Direct money access capability detected (payment gateways, crypto, banking).

• Direct money access detected (high risk: 1.00). The skill is explicitly a crypto exchange trading skill (Gate) and is designed to perform real financial actions: it drafts and executes spot buys/sells, USDT perpetual (futures) open/close/reverse/amend/cancel, margin borrowing, order/position management, and requires authenticated API keys (GATE_API_KEY/GATE_API_SECRET) or gate-cli config for private execution. These are specific exchange/transaction capabilities (crypto trading and order execution), not generic tooling, so it grants direct financial execution authority.