The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill directs users to clone a repository from an unverified GitHub account (https://github.com/YuhangChen1/Paper2All.git) and install unpinned dependencies via uv pip install -r requirements.txt.
[REMOTE_CODE_EXECUTION]: The skill instructions involve executing Python scripts (pipeline_all.py, pipeline_light.py) that are downloaded from a third-party remote source, allowing for the execution of arbitrary code from that repository.
[COMMAND_EXECUTION]: The installation documentation (references/installation.md) instructs the user to use sudo apt-get install for system dependencies like libreoffice and poppler-utils, requiring administrative privileges on the host system.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes untrusted LaTeX and PDF files provided by users.
Ingestion points: Files located in the input/ directory processed by the Paper2All pipeline.
Boundary markers: No specific boundary markers or instructions to ignore embedded commands within the processed papers were identified.
Capability inventory: The pipeline executes shell commands for document conversion (LibreOffice/Poppler) and makes external API calls (OpenAI/Google Search).
Sanitization: No evidence of input validation or sanitization of the research paper content is present.

paper-2-web