script-evaluator
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [Indirect Prompt Injection] (LOW): The skill is designed to ingest and process untrusted text data (full movie scripts or fragments). It lacks explicit boundary markers (e.g., XML tags or delimiters) to separate the untrusted input from the evaluation instructions, which could allow a script to contain hidden instructions that bias the report.
- Ingestion points: User-provided script content or fragments specified in
SKILL.md. - Boundary markers: Absent; the instructions do not specify how to wrap the input script to prevent instruction leakage.
- Capability inventory: Uses
ReadandWritetools to analyze files and generate reports. - Sanitization: None detected; the skill directly processes the input content based on the provided evaluation framework.
Audit Metadata