data-transform
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 'Transform, clean, reshape, and preprocess data', which creates an ingestion point for untrusted data. Without boundary markers or sanitization logic evidenced in this file, malicious instructions embedded in processed data could potentially influence agent behavior.
- Ingestion points: External data sources processed by pandas.
- Boundary markers: Absent in metadata.
- Capability inventory: Data transformation via pandas/numpy.
- Sanitization: Not evidenced in the metadata file.
- [Implementation Missing] (SAFE): The SKILL.md file containing the actual instructions and logic was not provided for analysis, limiting the scope to metadata only.
Audit Metadata