competitor-content-analysis
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and analyze content from arbitrary external competitor websites, creating a surface for indirect prompt injection. A malicious site could include hidden text designed to override the agent's instructions.
- Ingestion points: Competitor URLs, sitemap files, and various web pages (blog, guides, comparison pages) fetched during Steps 2, 4, 6, and 7.
- Boundary markers: The instructions do not define boundary markers or explicitly instruct the agent to ignore instructions contained within the fetched content.
- Capability inventory: The skill utilizes web fetching (Playwright MCP) and web search tools to retrieve external data.
- Sanitization: There are no sanitization or validation steps mentioned to filter out potential malicious prompts from the retrieved content.
Audit Metadata