scroll-audit
Pass
Audited by Gen Agent Trust Hub on Oct 29, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from external websites or local files, creating a potential surface for indirect prompt injection.
- Ingestion points: External web content is fetched for analysis in the 'Fetch and analyse' procedure (SKILL.md) and the 'Detect the scroll experience' reference (references/detection.md).
- Boundary markers: The skill lacks explicit delimiters or instructions for the agent to treat fetched content strictly as data or to ignore any instructions found within the audited pages.
- Capability inventory: The agent can access external networks via browser tools, read local files, write report files (remediation-plan.md), and potentially perform code fixes via the cinematic-scroll skill.
- Sanitization: No sanitization or safety validation of the ingested content is prescribed.
Audit Metadata