competitor-profiling
Pass
Audited by Gen Agent Trust Hub on Nov 10, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted content from external websites, creating a surface for indirect prompt injection attacks.
- Ingestion points: Firecrawl (map, scrape, crawl) and WebFetch tools are used to retrieve content from arbitrary competitor URLs specified by the user (SKILL.md, references/tool-reference.md).
- Boundary markers: While SKILL.md (Section 5) instructs the agent to ignore embedded instructions, it does not define strict structural delimiters to isolate the untrusted data from the rest of the prompt context.
- Capability inventory: The agent is authorized to write synthesized profile files to the local filesystem (competitor-profiles/ directory) and execute research-oriented MCP tools (SKILL.md).
- Sanitization: The skill relies on the AI agent's adherence to a 'Facts Over Opinions' principle to identify malicious input rather than programmatically sanitizing the website content.
Audit Metadata