The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/generate-slides.py uses subprocess.check_call to automatically install the google-genai package using the system's Python executable. This is used to ensure the environment is ready for image generation.
[EXTERNAL_DOWNLOADS]: The skill retrieves external dependencies at runtime. It uses pip to download the google-genai package from PyPI and employs npx to run Node.js tools, which may download packages like pdfjs-dist, canvas, and pptxgenjs from the NPM registry.
[REMOTE_CODE_EXECUTION]: Runtime installation and execution of external packages from PyPI and NPM are present. However, the packages identified are well-known libraries from trusted sources (Google, PDF.js contributors, etc.) required for the skill's primary purpose of PDF processing and image generation.
[PROMPT_INJECTION]: The skill is subject to indirect prompt injection risks (Category 8) as it ingests untrusted data from academic papers and user-provided content.
Ingestion points: Untrusted data enters the agent context via source-paper.pdf or text provided during the analyze content step.
Boundary markers: The skill does not currently implement explicit delimiters or instructions for the agent to ignore potentially malicious directions embedded within the source PDF text.
Capability inventory: The skill possesses capabilities for file system access, shell command execution (via Python and Node scripts), and network operations via the Gemini API.
Sanitization: There is no evidence of sanitization or filtering of the input text before it is interpolated into prompts for outline generation and image creation.

paper-slide-deck