skill-safety-scanner
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill requires the agent to execute a local Python script (
scripts/safety_scan.py) to perform its analysis. The script uses standard libraries and does not invoke shells unsafely or execute the files it scans. - [INDIRECT_PROMPT_INJECTION]: The skill possesses a data ingestion surface as it reads and analyzes local files provided in a directory path. It mitigates risks by using static regex-based pattern matching and does not dynamically execute or interpret the scanned content as instructions. Findings related to secrets are automatically redacted in the tool's output snippets.
- [SAFE]: The skill does not contain any obfuscated code, hidden network exfiltration logic, or external dependencies. It follows security best practices by providing a tool for local validation prior to publication.
Audit Metadata