cohort-analysis

Pass

Audited by Gen Agent Trust Hub on Sep 15, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted external data in CSV, Excel, and JSON formats, which could be used to deliver malicious instructions to the agent context.
  • Ingestion points: Step 1 (Read and Validate Your Data) and Example 1/3.
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are specified in the prompt instructions.
  • Capability inventory: The skill can read data, generate Python analysis scripts, and create visualizations.
  • Sanitization: No explicit sanitization or validation of data content for injection patterns is described.
  • [DYNAMIC_EXECUTION]: The skill generates Python scripts for data analysis based on user-provided metrics.
  • Evidence: Step 2 and the 'Key Capabilities' section explicitly mention generating Python analysis scripts using the pandas and numpy libraries.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 15, 2026, 12:02 PM