idor-hunting

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 1.00). The prompt explicitly instructs pasting and embedding session cookies (and shows an example session token) in PoCs and Autorize configuration, which requires handling/outputting secret tokens verbatim and is high-risk.

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document contains step-by-step, actionable instructions to perform unauthorized access (IDOR), data exfiltration, privilege escalation, and even unauthorized financial actions — highly dual-use but clearly enabling malicious activity if abused.

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

• Direct money access detected (high risk: 1.00). The skill includes explicit API examples that perform money movement or refunds: a POST /api/v3/transfer with from_account/to_account/amount (attempting unauthorized money movement) and a POST /api/v3/admin/billing/refund with an "amount" field. These are direct financial transaction actions, so the skill grants direct financial execution capability.