llm-crawler-access-check
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes robots.txt files from external, user-specified domains, which serves as an ingestion point for untrusted data. This is a standard functionality for this type of utility.
- Ingestion points: Content fetched from remote robots.txt files as specified in the Fetch section of SKILL.md.
- Boundary markers: Not explicitly defined; the skill relies on the agent to parse rules according to a provided logic table.
- Capability inventory: Restricted to network read access for a single public file and reporting results; no shell execution or file-write capabilities are requested.
- Sanitization: No explicit text sanitization is performed on the ingested content.
- [EXTERNAL_DOWNLOADS]: The skill performs network requests to fetch robots.txt files from remote domains. This is the primary intended function of the diagnostic tool and is limited to public metadata files.
Audit Metadata