clean
Pass
Audited by Gen Agent Trust Hub on Jul 27, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes untrusted data files which presents an indirect prompt injection surface.
- Ingestion points: Data files are read from the
input/<dataset>/directory as specified in SKILL.md. - Boundary markers: There are no explicit instructions or delimiters used to warn the agent to ignore potential instructions embedded within the CSV, Excel, or JSON data content.
- Capability inventory: The skill utilizes the
Read,Write,Bash,Glob, andGreptools to manage directories and execute Python-based data transformations. - Sanitization: While the skill standardizes column names using regex to remove non-alphanumeric characters, the actual cell content of the datasets is not sanitized for potential downstream prompt injection attacks.
Audit Metadata