geo-llmstxt
Pass
Audited by Gen Agent Trust Hub on Feb 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill fetches and processes untrusted content from external websites using the
WebFetchtool to perform its primary function of analyzing and generatingllms.txtfiles. - Ingestion points: The agent retrieves data from
[domain]/llms.txt,[domain]/llms-full.txt,[domain]/sitemap.xml, and various HTML pages across the target domain (Steps 1 and 3 in Generation/Analysis modes). - Boundary markers: The instructions lack explicit delimiters or warnings to the agent to treat the fetched content strictly as data, which may allow embedded instructions in the web content to influence agent behavior.
- Capability inventory: The skill is granted access to powerful tools including
Bash,Write, andWebFetch. - Sanitization: There is no logic provided to sanitize or filter out instructional content (e.g., HTML comments or hidden text) from the fetched data before it is processed by the agent.
Audit Metadata