mapping-mitre-attack-techniques

CRITICAL E004: Prompt injection detected in skill instructions.

• Potential prompt injection detected (high risk: 0.80). The prompt explicitly adds "Exploitation" and "Post-exploitation" steps that instruct offensive actions inconsistent with the skill's advertised analytical/mapping purpose, i.e., a deceptive scope expansion beyond the stated functionality.

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

• Third-party content exposure detected (high risk: 0.85). The required workflow downloads and parses the MITRE ATT&CK STIX bundle from a public GitHub URL at runtime (curl to enterprise-attack.json / mitreattack-python reads it), which is outsider-authored free text/structured prose from an external public source.