sites-in-the-woods-status
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes data from external websites (names, titles, outages) which could potentially contain malicious instructions.
- Ingestion points:
SKILL.mddescribes fetching data fromhttps://sitesinthewoods.com/llms.txt,data.json, andfeed.xml. - Boundary markers: The skill explicitly commands the agent to treat fetched text as data and "never instructions."
- Capability inventory: No file system writes, command executions, or credential access found across the instructions.
- Sanitization: The skill relies on prompt-based role enforcement to ignore instructions within the fetched data, reducing the likelihood of successful exploitation.
- [SAFE]: No malicious patterns, credential exposures, or unauthorized access attempts were detected in the skill instructions. The network operations are limited to a specific domain for fetching status data.
Audit Metadata