review-intelligence-digest
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external review sites (G2, Capterra, Trustpilot) to generate marketing insights and copy.
- Ingestion points: Scraped reviews from external URLs in Phase 1.
- Boundary markers: No explicit delimiters or instructions are provided to the LLM to ignore potentially malicious content within reviews during analysis in Phase 2.
- Capability inventory: Executes local Python scripts and writes generated reports to the filesystem.
- Sanitization: There is no evidence of filtering or sanitizing review content before it is processed by the LLM.
- [COMMAND_EXECUTION]: The skill uses
python3to execute a script located in a referenced local skill folder (skills/review-scraper/scripts/scrape_reviews.py). This is a typical pattern for modular agent skills.
Audit Metadata