yao-positioning-skill

Pass

Audited by Gen Agent Trust Hub on Jul 17, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill includes a vendored version of the Apache ECharts visualization library in assets/echarts.min.js. This is a well-known, trusted component used for rendering interactive charts within the offline HTML reports.
  • [COMMAND_EXECUTION]: The workflow relies on local Python scripts (scripts/render-report.py, scripts/validate-intake.py, etc.) to validate input data and render report artifacts. These scripts use the Python standard library and do not exhibit dangerous command execution or subprocess spawning patterns.
  • [DATA_EXFILTRATION]: The skill includes a security feature in scripts/sanitize-package.py that explicitly scans for and redacts local absolute file paths (such as macOS /Users/ or Windows \Users\ patterns) from generated reports to prevent accidental data disclosure when sharing report bundles.
  • [SAFE]: The skill defines a formal permission policy in security/permission_policy.json, scoping its file-write capabilities to the generation of specific report artifacts. It implements high-quality sanitization, such as safe_json_for_script and HTML escaping, to protect against potential cross-site scripting (XSS) when processing user-provided content.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 17, 2026, 03:48 AM
Security Audit — agent-trust-hub — yao-positioning-skill