apify-sdk-patterns
Pass
Audited by Gen Agent Trust Hub on Sep 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill includes patterns for web scraping that ingest untrusted content from external websites into the agent's context.
- Ingestion points: The crawler request handlers in SKILL.md and references/patterns.md are designed to extract data from third-party URLs.
- Boundary markers: The provided code templates do not include specific delimiters or 'ignore' instructions to separate untrusted scraped data from agent instructions.
- Capability inventory: The skill demonstrates capabilities for writing to storage using Actor.pushData and Actor.setValue, as well as making authenticated network requests through the ApifyClient.
- Sanitization: The snippets use simple string methods like trim() but do not implement formal sanitization or validation for the content being processed.
Audit Metadata