similar-judge

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The utility invokes git diff and awk to process text differences.
  • The script implements security best practices by using double-quotes for variables ("$target") and the -- separator to prevent argument injection via malicious filenames.
  • [SAFE]: The skill reads local files to perform text analysis.
  • Access is limited to the file paths provided by the agent as arguments. The content is processed locally to generate similarity ratios and is not transmitted to external domains.
  • [SAFE]: The external repository https://github.com/gitByEOS/open-part-skills is owned by the skill author gitByEOS and is used for metadata purposes.
  • [SAFE]: The Python logic used for similarity calculation (difflib) is implemented using standard library modules and does not introduce external dependencies or dynamic execution risks.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 10:02 AM
Security Audit — agent-trust-hub — similar-judge