The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill's primary function is to process external PDF files from URLs or local paths. The extracted text from these files is used as the primary context for generating study materials and runnable code. A malicious PDF could contain instructions designed to manipulate the agent's behavior or output.
Ingestion points: The skill downloads and parses PDF files provided via user input (URL or local path) in Step 1.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands are present when the agent processes the extracted text.
Capability inventory: The skill possesses the Bash, Write, and Edit tools, enabling it to create scripts and manage files.
Sanitization: The skill does not sanitize or filter the extracted PDF content before using it as a source for material generation.
[EXTERNAL_DOWNLOADS]: The download-pdf.cjs script enables the agent to download files from any HTTP/HTTPS URL provided by the user. While the script performs basic validation (e.g., checking for the application/pdf content-type), it allows the agent to fetch data from arbitrary remote servers.
[COMMAND_EXECUTION]: The skill uses the Bash tool to perform setup tasks (installing npm and pip packages), execute extraction scripts, and perform file system operations like creating directories and copying artifacts.
[DYNAMIC_EXECUTION]: The agent is specifically instructed to generate runnable code demonstrations (Python scripts or Jupyter Notebooks) and interactive HTML files based on the analyzed paper's content. This represents a dynamic code generation pattern where the resulting code's logic is influenced by untrusted external input.

study