matlab-test-execution

Pass

Audited by Gen Agent Trust Hub on Mar 22, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill facilitates legitimate MATLAB software testing and quality assurance workflows. No malicious obfuscation, persistence, or data exfiltration patterns were identified.- [EXTERNAL_DOWNLOADS]: The skill references official MathWorks GitHub Actions (matlab-actions/setup-matlab, matlab-actions/run-build) and Docker images (mathworks/matlab) for CI/CD integration. These are well-known resources provided by the platform vendor for MATLAB automation.- [COMMAND_EXECUTION]: The skill instructs the agent to execute MATLAB commands using the evaluate_matlab_code tool or the matlab -batch command. This execution is necessary for the skill's intended purpose of running unit tests and collecting coverage data.- [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection as it processes and displays test diagnostic reports and coverage data which are generated from external code. Ingestion points: Test failure reports in SKILL.md and coverage result objects in scripts/printCoverageGaps.m. Boundary markers: No explicit boundary markers or ignore-instructions are used when displaying diagnostics. Capability inventory: MATLAB code execution via MCP or shell. Sanitization: The skill does not perform sanitization on diagnostic strings before they are processed by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 22, 2026, 09:51 AM