chart-vision
Pass
Audited by Gen Agent Trust Hub on May 10, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill documentation and provided code examples are consistent with its stated purpose of chart vision and trading analysis. No evidence of obfuscation, exfiltration, or unauthorized command execution was found.
- [EXTERNAL_DOWNLOADS]: The skill's technical stack relies on established and trusted libraries such as Matplotlib, OpenCV, and Scikit-learn. While specific version numbers in the documentation (e.g., numpy 2.4, scipy 1.17) appear to be placeholders or future-dated, the package names themselves are legitimate.
- [COMMAND_EXECUTION]: Code snippets demonstrate routine file operations including reading CSV data and saving image files locally. These operations are scoped to the project's local directory and do not involve suspicious system-level commands.
- [PROMPT_INJECTION]: The skill processes external data sources (OHLCV CSV files and chart screenshots). While this represents a surface for indirect prompt injection, the skill does not include instructions to bypass safety guidelines or reveal system prompts. No boundary markers are currently present in the examples to isolate untrusted data.
Audit Metadata