chart-image
Pass
Audited by Gen Agent Trust Hub on Nov 5, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses a Node.js script (chart.mjs) that is executed via the command line to process data and generate image files. It includes functionality to read from local files and write results to the file system based on user-supplied arguments.
- [EXTERNAL_DOWNLOADS]: The skill instructs the user to install dependencies (vega, vega-lite, sharp) from the official NPM registry during the initial setup phase. These are well-known and trusted libraries for data visualization and image processing.
- [INDIRECT_PROMPT_INJECTION]: The skill processes external data (JSON, CSV, or shorthand text), which creates a potential surface for indirect prompt injection if the agent were to interpret the visual content or data structure as new instructions. 1. Ingestion points: The chart.mjs script ingests data via --data, --csv-file, --spec, and standard input. 2. Boundary markers: The skill does not implement explicit boundary markers or warnings to the agent regarding instructions embedded in input data. 3. Capability inventory: The skill has the capability to read local files (readFileSync) and write to the local file system (writeFileSync). 4. Sanitization: Input is parsed using standard methods like JSON.parse and a dedicated CSV parser, which limits the risk by treating the input as structured data rather than executable code.
Audit Metadata