skills/mukul975/anthropic-cybersecurity-skills/hunting-for-unusual-service-installations/Gen Agent Trust Hub
hunting-for-unusual-service-installations
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes external Windows Event Log (.evtx) files, which could potentially contain malicious payloads designed to influence the processing logic.
- Ingestion points: The scripts/agent.py script reads .evtx files provided by the user.
- Boundary markers: No specific delimiters or instructions are used to separate log content from processing instructions.
- Capability inventory: The skill only performs read-only analysis and prints results to stdout; it has no capabilities for network communication, file system modification, or command execution.
- Sanitization: Event data is printed as JSON, which prevents many forms of basic injection, though the raw content is not filtered.
- [SAFE]: The skill operates as intended for forensic analysis and does not contain any malicious code, obfuscation, or data exfiltration mechanisms.
Audit Metadata