writing-reader-feedback
Pass
Audited by Gen Agent Trust Hub on Aug 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from multiple sources which could contain malicious instructions designed to influence agent behavior.
- Ingestion points: Workflow Step 1 in SKILL.md identifies direct text, local file paths, and webpage URLs as input sources.
- Boundary markers: The instructions lack explicit delimiters or warnings to ignore instructions embedded within the article text being analyzed.
- Capability inventory: The skill utilizes file system access and network fetching tools (url-reader, WebFetch).
- Sanitization: No content validation or escaping mechanisms are implemented for the ingested text.
Audit Metadata