matlab-exclude-files

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the 'evaluate_matlab_code' tool to perform file system operations, including recursive globbing of directory contents and reading or writing 'toolbox.ignore' and 'package.ignore' files. These actions are necessary for the skill's core purpose of identifying and managing excluded files.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface by processing untrusted data from the user's file system, specifically filenames and the content of existing ignore files.\n
  • Ingestion points: File paths and names are collected via recursive globbing in the workflow, and existing ignore files are read from the project directory.\n
  • Boundary markers: No explicit delimiters or instructions are provided to the agent to distinguish external data (filenames) from its primary instructions.\n
  • Capability inventory: The skill utilizes the 'evaluate_matlab_code' tool for file system interaction and potentially code execution within the MATLAB environment.\n
  • Sanitization: There is no evidence of filtering, escaping, or validation of the data ingested from the file system.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 09:27 AM
Security Audit — agent-trust-hub — matlab-exclude-files