decision-matrix
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFE
Full Analysis
- [SAFE] (SAFE): No malicious code or commands detected. The skill is purely instructional markdown and does not contain any executable scripts or shell commands.- [NO_CODE] (SAFE): The skill consists entirely of markdown instructions and does not include any scripts (Python, JS, etc.) or configuration files that trigger code execution.- [DATA_EXPOSURE] (SAFE): No access to sensitive file paths (e.g., ~/.ssh, ~/.aws) or environment variables was found. The skill processes user-provided data for decision-making purposes only.- [EXTERNAL_DOWNLOADS] (SAFE): No external URLs, remote package installations (pip, npm), or remote script executions (curl | bash) are present.- [INDIRECT_PROMPT_INJECTION] (SAFE): While the skill ingests user-provided alternatives and criteria, it lacks executable capabilities (file writes, network requests, subprocesses) that could be exploited through malicious data input.
Audit Metadata