site-content-catalog
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external websites (HTML, XML, RSS) and the Apify API, creating a surface for indirect prompt injection where malicious instructions embedded in web content could influence the AI agent's behavior during subsequent processing.
- Ingestion points:
scripts/catalog_content.pyfetches data from target domains viarequests.getand from Apify viarequests.post. - Boundary markers: None; extracted titles and metadata are placed directly into JSON and Markdown structures without delimiters or safety warnings for the agent.
- Capability inventory: The script performs network operations (
requests) and file system writes (open(..., 'w')) for catalog output. - Sanitization: Content is processed via XML parsing and regex for structural extraction, but no sanitization is performed to neutralize potential prompt instructions within the extracted text.
Audit Metadata