matlab-analyze-data

Pass

Audited by Gen Agent Trust Hub on Oct 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill provides high-quality, idiomatic guidance for MATLAB data analysis workflows using tables, timetables, and arrays.\n- [SAFE]: The documentation includes security and stability recommendations, such as avoiding the use of eval for dynamic variable access in 'references/tables-and-timetables.md' and delegating credential management to specialized secure tools.\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it generates code to process external data (tables, sensor data, gridded data) and executes it via the evaluate_matlab_code tool.\n
  • Ingestion points: Tabular and array data in workspace variables, which may originate from external files or sensor logs (specified in 'SKILL.md').\n
  • Boundary markers: The skill does not specify boundary markers or instructions to ignore embedded commands within the processed data values.\n
  • Capability inventory: The 'evaluate_matlab_code' tool is used to execute generated MATLAB analysis code (referenced in 'manifest.yaml').\n
  • Sanitization: The instructions recommend data cleaning via standardizeMissing and strictly forbid the use of eval in 'references/tables-and-timetables.md', reducing the risk of accidental execution of data-resident instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 9, 2026, 07:12 PM