matlab-debug-code
Pass
Audited by Gen Agent Trust Hub on Aug 14, 2026
Risk Level: SAFEPROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data, specifically MATLAB code snippets, error messages, and stack traces provided by users or generated at runtime. This data is ingested and analyzed using tools with execution capabilities, creating a potential surface for indirect instructions to influence agent behavior.
- Ingestion points: The agent processes user code and MATLAB runtime outputs as described in the investigation workflow.
- Boundary markers: There are no instructions for using specific delimiters or 'ignore' directives to isolate external content from instructions.
- Capability inventory: The skill uses tools like evaluate_matlab_code and run_matlab_file which execute code in the environment.
- Sanitization: The workflow does not mention sanitizing or validating error outputs before processing them.
Audit Metadata