The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python utility scripts (e.g., material_inventory.py, integrity_audit.py) to automate the processing of research documents. These scripts use the Python standard library and are designed for local file system operations related to the academic writing task.
[DATA_EXPOSURE]: The skill processes user-provided research materials from a specified directory. While it reads sensitive research data, it does so locally to create intermediate artifacts and the final LaTeX project. No network calls or exfiltration patterns were identified.
[PROMPT_INJECTION]: The instructions contain strong internal consistency and safety guardrails, such as requiring user confirmation for motivations and prohibiting the fabrication of experimental results. No attempts to bypass agent safety filters or override system instructions were found.

paper-spine-build