The Agent Skills Directory

[DYNAMIC_EXECUTION]: The file lib/dom-setup.js configures the jsdom environment with the runScripts: 'dangerously' option. This allows the execution of any JavaScript contained within the SVG or HTML content being rendered for the infographic. Additionally, lib/exporter.js launches puppeteer with the --no-sandbox and --disable-setuid-sandbox flags, which reduces the security isolation of the headless browser process.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data and assets, creating a vulnerability surface for indirect instructions.
Ingestion points: The skill ingests user-provided JSON data via the --data flag in generate.js and loads external image files from a configurable directory in lib/illustrations.js.
Boundary markers: The skill does not implement clear boundary markers or instructions to ignore potential injections within the ingested data during interpolation.
Capability inventory: The skill is granted access to the Bash, Write, and Read tools. It also launches external processes, including a headless browser and a script-enabled virtual DOM.
Sanitization: The sanitizeSVG function in lib/illustrations.js only removes XML declarations and comments. It fails to strip dangerous elements such as <script> tags or event handlers. Combined with the dangerous jsdom configuration, this creates a path for malicious code execution within the rendering environment if a poisoned SVG file is provided.
[COMMAND_EXECUTION]: The skill utilizes the Bash tool to execute the node generate.js command. This process involves passing user-influenced data through command-line arguments, which is a standard part of its workflow but increases the risk profile if inputs are not strictly validated.

generating-infographics