The Agent Skills Directory

[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface (Category 8) by processing untrusted data from external URLs and user-provided text. * Ingestion points: references/mode-infograph.md (Step 2) and SKILL.md describe fetching content from URLs and files. * Boundary markers: The skill does not define specific delimiters or instructions to ignore embedded commands within the processed data. * Capability inventory: The skill uses assets/capture.js to execute a Playwright-controlled Chromium browser to render content. * Sanitization: There is no evidence of sanitization or escaping of the content before it is interpolated into HTML templates (e.g., {{BODY_HTML}}).
[COMMAND_EXECUTION]: The skill relies on the execution of local Node.js scripts to perform its core functions. * Evidence: SKILL.md and associated mode documents provide instructions to execute node ~/.claude/skills/ljg-card/assets/capture.js to generate the PNG output.
[EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the Playwright library and its associated browser binaries from the NPM registry. * Evidence: package.json includes playwright as a dependency, and SKILL.md provides setup commands using npm and npx.

ljg-card