The Agent Skills Directory

[COMMAND_EXECUTION] (SAFE): The validate_format.py script uses subprocess.run to call external utilities like pdfinfo and pdffonts. The implementation is secure as it passes arguments as a list and does not use shell=True, preventing command injection vulnerabilities.
[DATA_EXFILTRATION] (SAFE): No network-related code (e.g., requests, urllib) or hardcoded credentials were found. The scripts operate entirely on local files within the skill's directory or user-specified paths.
[REMOTE_CODE_EXECUTION] (SAFE): The scripts do not download or execute code from external sources. The logic is restricted to string manipulation for LaTeX templates and metadata extraction from PDFs.
[PROMPT_INJECTION] (SAFE): All provided markdown files consist of benign examples of scientific abstracts and introductions. There are no instructions attempting to override agent behavior or bypass safety filters.

venue-templates