competitor-profiling

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external competitor websites via scraping tools.
  • Ingestion points: Third-party website content is fetched using firecrawl_scrape and firecrawl_search as described in Phase 1 of the research process.
  • Boundary markers: The skill includes a dedicated 'Untrusted Input' section in SKILL.md that explicitly directs the agent to treat competitor pages as data to analyze rather than instructions to follow, warning about hidden HTML directives.
  • Capability inventory: The skill uses network tools (firecrawl, dataforseo) for information retrieval and local file system operations to save generated profiles and raw research data to the competitor-profiles/ directory.
  • Sanitization: The skill relies on natural language instructions to the agent to disregard any embedded instructions within the scraped data and to note such attempts in the final profile.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes well-known external services (Firecrawl for scraping and DataForSEO for market metrics) to gather competitive intelligence. These are established tools for market research and their use is consistent with the skill's stated purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 05:01 AM