The Agent Skills Directory

[COMMAND_EXECUTION]: The skill's workflow in SKILL.md requires executing a bash script (scripts/convert.sh) that manages the conversion pipeline, including directory management, file movement, and tool invocation.
[EXTERNAL_DOWNLOADS]: The scripts/convert.sh script executes npx with the --yes flag to automatically download and run external packages from the NPM registry:
md-to-pdf: Used for the Markdown to PDF conversion process.
@mermaid-js/mermaid-cli: Used to render Mermaid diagrams into PNG images.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it processes user-provided Markdown files without implementing boundary markers or instructions to ignore embedded commands.
Ingestion points: The agent reads local .md files as specified in SKILL.md and processes their contents via preprocess.cjs.
Boundary markers: No delimiters or warnings are used to separate the content of the files from the agent's internal instructions.
Capability inventory: The skill possesses capabilities for shell command execution (bash), package execution (npx), and file system modification.
Sanitization: No validation or filtering is applied to the Markdown content to prevent instructions embedded in the text from influencing the agent's behavior.

md-to-pdf