dineway-site-analysis
Pass
Audited by Gen Agent Trust Hub on Sep 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDATA_EXFILTRATION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill crawls external websites and processes their content, which could contain malicious instructions designed to influence the agent.\n
- Ingestion points: The agent crawls public pages (SKILL.md, Step 2) and competitive evidence (SKILL.md, Step 6).\n
- Boundary markers: No specific boundary markers or instructions to ignore embedded commands are defined.\n
- Capability inventory: The skill uses Browser Use, writes to the local file system in the .dineway/ directory, and executes Content Pipeline MCP tools.\n
- Sanitization: The skill lacks explicit sanitization logic for data retrieved from external sources.\n- [DATA_EXFILTRATION]: The skill performs network operations to crawl websites and access third-party SEO services like Firecrawl and DataForSEO. These operations are within the stated scope of site analysis but target non-whitelisted domains.
Audit Metadata