The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses the subprocess module to execute external system utilities for rendering and verifying diagrams. Specifically, scripts/render_svg.py, scripts/verify_render_font_equivalence.py, and the unit tests in scripts/tests/ invoke rsvg-convert and magick (ImageMagick) to process SVG files and compare image output.
[DYNAMIC_EXECUTION]: The verification script scripts/verify_render_font_equivalence.py and the generator tests in scripts/tests/test_svg_producer_contract.py dynamically execute other Python scripts within the skill using sys.executable to perform regression testing and property validation.
[INDIRECT_PROMPT_INJECTION]: The skill processes external, user-provided Markdown documents, which introduces a surface for indirect prompt injection attacks.
Ingestion points: extract_svgs.py and scan_consistency.py read and parse Markdown files specified by the user to extract or validate inline SVG blocks.
Boundary markers: The scripts do not utilize specific delimiters or "ignore instructions" warnings when ingesting the contents of these documents.
Capability inventory: The skill possesses the capability to write files to the local file system and execute shell commands via subprocess calls.
Sanitization: Content extracted from Markdown files undergoes minimal sanitization before being processed by the SVG extraction logic.
[EXTERNAL_DOWNLOADS]: The scripts/svg2png.js utility relies on the puppeteer Node.js package for browser-based rendering. The skill's documentation in SKILL.md identifies this as a required external dependency and provides instructions for manual installation via the standard npm registry.

svg-book-illustrator