The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill automates the installation of necessary dependencies using standard package managers. The script scripts/check_deps.sh invokes pip3 install for Python packages (reportlab, weasyprint, markdown, pygments) and brew install for system libraries (pango, cairo, gdk-pixbuf, libffi) on macOS.
[COMMAND_EXECUTION]: The skill uses the Bash tool to orchestrate the conversion process. It executes the dependency checker, the main Python conversion script (scripts/md_to_pdf.py), and various inline Python snippets for configuration management and version extraction.
[INDIRECT_PROMPT_INJECTION]: The skill includes a 'Convert+Prompt' mode that reads an input Markdown file and applies LLM-based transformations (e.g., restructuring or redacting content) before generating the PDF. This processes untrusted external data within the agent's reasoning loop.
Ingestion points: The content of the user-provided Markdown file is read using the Read tool in SKILL.md Step 3.
Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore potentially malicious instructions embedded within the Markdown source during the preprocessing phase.
Capability inventory: The skill has access to Bash, Read, Write, Edit, Glob, and Grep tools, which could be abused if the agent follows instructions found in the source file.
Sanitization: No sanitization or filtering of the Markdown content is performed before it is processed by the LLM.
[DYNAMIC_EXECUTION]: In the STYLES mode section of SKILL.md, the skill generates and executes a multi-line Python script using a shell heredoc (`python3
<<'PY' ... PY`). This script is used to perform a deep merge of style configurations and metadata into the project's local JSON config file.

md-to-pdf