senior-data-scientist
Pass
Audited by Gen Agent Trust Hub on Apr 12, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill exhibits an attack surface for indirect prompt injection via data processing and formula interpolation.
- Ingestion points: Statistical functions in
SKILL.md(e.g.,analyze_experiment,diff_in_diff) ingest data and column names from potentially untrusted datasets. - Boundary markers: The instructions do not specify any delimiters or safety markers for ingested data.
- Capability inventory: The agent context includes high-privilege tools such as
docker,kubectl, andhelmas listed inSKILL.md. - Sanitization: Column names are interpolated into
statsmodelsformulas using f-strings (e.g., 'f{outcome} ~ {treatment_col}') without prior validation.
Audit Metadata