market-competitors
Pass
Audited by Gen Agent Trust Hub on Sep 21, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes content from third-party websites and review platforms like G2 and Capterra. Ingestion points: Fetches target and competitor URLs as specified in SKILL.md. Boundary markers: None identified in the instructions. Capability inventory: Uses WebFetch for network reads, writes COMPETITOR-REPORT.md, and runs a local collection script. Sanitization: No mention of filtering or sanitizing fetched content.
- [COMMAND_EXECUTION]: The skill runs a Python script with user-supplied URL arguments. Evidence: python scripts/competitor_scanner.py --url [competitor-url] --output json in SKILL.md. Risk: Potential command injection if the execution environment does not sanitize the [competitor-url] variable.
Audit Metadata