skill-guard

Pass

Audited by Gen Agent Trust Hub on Mar 11, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's instructions and configuration are entirely consistent with its stated purpose as a defensive security monitor. It contains no malicious code, unauthorized exfiltration logic, or obfuscation. The author 'useclawpro' is identified as a vendor resource associated with the platform's security tooling.\n- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection (Category 8) because it is designed to analyze data generated by third-party skills, which may be untrusted.\n
  • Ingestion points: The skill monitors file paths, network logs, and shell commands produced during the execution of other skills.\n
  • Boundary markers: While it provides a structured alert format, it lacks specific delimiters or instructions to ignore embedded commands within the data it monitors.\n
  • Capability inventory: The skill has fileRead permissions but is restricted by its own rules to act solely as a read-only monitor.\n
  • Sanitization: No explicit logic is defined to sanitize or validate the content of the behaviors it observes.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 11, 2026, 01:32 AM