The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data from user-provided .zip files and conference submission URLs.
Ingestion points: Receives and extracts .zip archives in Step 1 and fetches content from user-supplied conference links in Step 6 to extract submission requirements.
Boundary markers: The instructions lack explicit delimiters or safety prompts that would prevent the agent from following instructions embedded within the processed LaTeX files or fetched web content.
Capability inventory: The skill has access to shell commands (unzip, find, grep, mkdir, cp, rm) and the ability to write files to the local file system.
Sanitization: There is no evidence of sanitization for filenames or web data before they are used in commands or interpolated into generated files.
[COMMAND_EXECUTION]: The skill instructions rely on several shell commands to manipulate the file system.
Evidence: Uses bash commands like unzip, find, grep, mkdir, cp, and rm to restructure the LaTeX project and manage temporary files.
[EXTERNAL_DOWNLOADS]: The skill performs network operations to retrieve conference information.
Evidence: Uses a web fetching capability to access external URLs provided by the user for extracting submission requirements.

latex-conference-template-organizer