implementing-siem-use-cases-for-detection

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

• Third-party content exposure detected (high risk: 0.70). The skill explicitly fetches and uses open third‑party ATT&CK data via attackcti in scripts/agent.py (get_attack_coverage) and the SKILL.md workflow instructs pulling/executing an Atomic Red Team install script from raw.githubusercontent.com (Step 4), so the agent is expected to read/ingest public external content that can materially influence coverage calculations and testing/deployment decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill includes a runtime PowerShell command that downloads and executes remote code via IEX from https://raw.githubusercontent.com/redcanaryco/invoke-atomicredteam/master/install-atomicredteam.ps1 (Install-AtomicRedTeam -getAtomics), which directly executes fetched code as part of testing and thus is a risky external dependency.