The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes shell commands to manage files (mkdir, cp) and perform format conversions using headless browser operations (google-chrome --headless) and office suite utilities (libreoffice --headless).
[COMMAND_EXECUTION]: It instructs the agent to execute a local Python script (scripts/generate_schematic.py) to create visual components for the posters based on structured prompt requirements.
[EXTERNAL_DOWNLOADS]: The documentation identifies python-pptx as a dependency for direct PPTX generation and mentions system-level dependencies for PDF rendering.
[PROMPT_INJECTION]: The skill processes user-provided data to populate poster content, creating a surface for indirect prompt injection.
Ingestion points: Poster metadata (title, authors) and section content are used to fill an HTML template and generate visualization prompts in SKILL.md.
Boundary markers: The instructions do not define specific delimiters for separating user content from the HTML structure or shell command arguments.
Capability inventory: The skill is configured with Bash, Write, Read, and Edit tools, allowing it to execute scripts and manipulate the file system.
Sanitization: No explicit validation or escaping of user input is specified before it is integrated into the poster templates or figure generation prompts.

pptx-posters