python-strategy

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

• Third-party content exposure detected (high risk: 0.90). The scanner explicitly calls mcporter_call("leaderboard_get_markets") and mcporter_call("market_get_asset_data") (see get_sm_map() and build_thesis() in scripts/python-scanner.py), ingesting public market feeds and leaderboard/trader-derived data (untrusted, user-generated) which the agent parses and uses to score, block, and drive entry/execution decisions.

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 trading agent: it targets "crypto only" perpetuals (top 50 HL perps), requires a WALLET_ADDRESS in runtime config, defines leverage, margin, position-sizing, hard drawdown breakers, and references a function "strategy_get_asset_trading_limits" and runtime/scripts (python-scanner.py) intended to run on a cron. These elements are specific to placing and managing market positions (orders) and interacting with crypto trading infrastructure/wallets — not generic tooling. Under the CORE RULE, this is specifically designed for crypto/market order execution and thus grants direct financial execution capability.