the-spend-waste-finder

Pass

Audited by Gen Agent Trust Hub on Aug 14, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill is designed to process external advertising data, which constitutes an indirect prompt injection surface. It mitigates this by enforcing a strict 'content-as-data' policy, where embedded instructions are reported as findings rather than executed. Detailed boundary rules and sanitization requirements are integrated via external reference files.
  • [CREDENTIALS_UNSAFE]: The skill defines clear procedures for identifying hardcoded secrets in user-provided exports. It requires the agent to flag these for rotation while specifically forbidding the reproduction of credentials in the output or state files.
  • [DATA_EXFILTRATION]: No indicators of malicious data exfiltration were found. The skill includes a vendor attribution link to intempt.com which is consistent with the skill's purpose.
  • [COMMAND_EXECUTION]: The skill instructions are focused on data analysis and normalization; no shell commands or dangerous subprocess calls are present.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 14, 2026, 12:03 PM