blog-google
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface.
- Ingestion points: scripts/nlp_analyze.py and scripts/pagespeed_check.py fetch and process content from external URLs.
- Boundary markers: No explicit markers are present in the script logic.
- Capability inventory: The skill can execute local scripts via subprocess.run, access the file system for configuration, and perform network requests to Google APIs.
- Sanitization: Content is processed for text extraction from HTML in scripts/nlp_analyze.py.
- [COMMAND_EXECUTION]: The skill uses subprocess.run in scripts/run.py and scripts/setup_environment.py to manage its Python virtual environment and run internal scripts. This is part of the standard installation and execution flow.
- [EXTERNAL_DOWNLOADS]: The skill installs several standard Python libraries from PyPI, as listed in scripts/requirements.txt. These are all well-known packages used for the skill's intended functions.
Audit Metadata