scrapy-extra
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill facilitates the creation and execution of web spiders that ingest untrusted data from arbitrary external websites. This content could potentially contain malicious instructions designed to influence the AI agent's extraction logic or debugging process.
- Ingestion points: Untrusted data enters the agent context through
HttpResponseandBrowserResponsedependencies inSKILL.md,scrapy.md, andweb-poet.md, accessed via CSS and XPath selectors. - Boundary markers: The instructions do not define explicit boundary markers or instruction-ignore warnings to wrap scraped content during processing.
- Capability inventory: The skill enables network operations via
uv run scrapy crawland writes extracted data to the local file system (e.g.,items.jsonl). - Sanitization: The provided templates and instructions do not include explicit steps for sanitizing or escaping raw HTML content before it is analyzed by the agent.
Audit Metadata