pentest-recon-attack-surface
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources, providing a surface for indirect prompt injection attacks that could influence agent behavior.\n
- Ingestion points: The workflow involves capturing network traffic, XHR/fetch requests, and page content via tools like
httpx_probe,playwright_authenticated_crawl, andkatana_crawldefined inreferences/tools.md.\n - Boundary markers: There are no instructions for the agent to use delimiters or to treat data from external scans and crawls as potentially malicious, which could lead to the agent executing instructions embedded in the target application's responses.\n
- Capability inventory: The skill possesses broad capabilities, including performing network scans and reading local source code using
semgrep_route_extractandripgrep_pattern.\n - Sanitization: The documentation in
references/workflows.mddoes not specify any sanitization or validation of external content before the agent analyzes it to generate endpoint inventories or prioritize vulnerabilities.
Audit Metadata