The Agent Skills Directory

[PROMPT_INJECTION]: The skill contains an indirect prompt injection surface because it is designed to ingest and process untrusted data from MATLAB error messages and stack traces.
Ingestion points: MATLAB runtime errors, syntax errors, and stack traces analyzed during the "Auto-Trigger on MATLAB Errors" workflow in SKILL.md.
Boundary markers: Not implemented; the agent is instructed to recognize error patterns directly from tool output without delimiters.
Capability inventory: The skill utilizes high-privilege tools including mcp__matlab__evaluate_matlab_code and mcp__matlab__run_matlab_file as specified in manifest.yaml.
Sanitization: There is no explicit sanitization or escaping of the ingested MATLAB error content before it is processed by the agent's logic.

matlab-debugging