bb-methodology

CRITICAL E006: Malicious code pattern detected in skill scripts.

• Malicious code pattern detected (high risk: 1.00). This document is an explicit offensive playbook that directs abuse techniques (data exfiltration, credential theft, evasion, RCE, web shells, CI/CD/metadata theft and supply‑chain attacks) and even advises hiding some findings, indicating clear malicious intent and high abuse risk.

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

• Third-party content exposure detected (low risk: 0.10). The skill’s workflow ingestion at runtime is directed at attacker-chosen “target.com” via recon/mapping/exploit tools (e.g., recon commands, JS/code download, GraphQL/CI-CD scanning), which entails the LLM ingesting outsider-authored free text from scraped/enumerated web/endpoint content rather than only first-party provider docs.