always-verify-numbers-and-links
Pass
Audited by Gen Agent Trust Hub on Oct 23, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues were detected. The skill functions as a documentation guide to improve the accuracy and safety of AI responses.
- [PROMPT_INJECTION]: No malicious prompt injection or bypass attempts were found. The provided prompt templates are defensive and designed to prevent hallucinations by requiring citations and step-by-step verification.
- [DATA_EXPOSURE]: No hardcoded credentials, sensitive file paths, or unauthorized network operations were identified. References to external sites like IRS.gov or GitHub are for illustrative verification purposes.
- [DYNAMIC_EXECUTION]: The skill includes a Python code snippet as an example for calculating a budget. This code is static, performs simple arithmetic, and does not execute untrusted input or perform dangerous operations.
- [INDIRECT_PROMPT_INJECTION]: The skill provides guidance on how to safely handle external data (e.g., financial reports) by using specific constraints like verbatim quoting and strict fallback messages ('NOT SPECIFIED IN SOURCE'), which serves to mitigate injection risks rather than create them.
Audit Metadata