competitor-monitor
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes content from external websites, which are untrusted sources that could contain malicious instructions designed to influence agent behavior. \n
- Ingestion points: Untrusted website data is fetched via the
fetch_pagefunction inscripts/main.pyusing therequestslibrary. \n - Boundary markers: The skill displays content previews and saved snapshot metadata without using explicit delimiters or "ignore instructions" warnings for the agent. \n
- Capability inventory: The skill possesses the ability to make arbitrary network requests and write files to the local filesystem (e.g., saving snapshots and metadata). \n
- Sanitization: The script uses
BeautifulSoupto extract text from HTML, which removes tags but does not sanitize the resulting text content for potential prompt injection patterns.
Audit Metadata