The Agent Skills Directory

[COMMAND_EXECUTION]: The skill automates document generation by executing system commands and scripts. It runs scripts/analyze_and_generate_source.py to perform project analysis and generates a build_all.sh shell script to invoke the xelatex compiler. Documentation also includes environment setup procedures involving package managers like brew or apt, which may include the use of sudo for system-level installations.
[PROMPT_INJECTION]: The skill presents an attack surface for indirect prompt injection through the ingestion of external data.
Ingestion points: The Python analysis script reads and extracts content from all source files within a user-specified project directory.
Boundary markers: Code content is encapsulated within LaTeX Verbatim environments in the generated output.
Capability inventory: The agent has the capability to execute Python scripts, create directories, and run shell commands/scripts.
Sanitization: The analysis script escapes specific LaTeX control sequences (like \end{Verbatim}) to prevent document compilation errors, but it does not sanitize content against potential agent instructions embedded in project comments or file metadata.

chinese-copyright-application