The Agent Skills Directory

[COMMAND_EXECUTION]: The skill invokes a local Python script (look_tongji.py) and the xelatex typesetting engine to generate documents. These commands are localized to the user's repository environment and serve the primary function of the skill.
[PROMPT_INJECTION]: The skill has an indirect prompt injection surface as it is designed to ingest and process course lecture notes. Ingestion points: Lecture notes read by the agent (SKILL.md). Boundary markers: None explicitly defined in the provided workflow to isolate embedded instructions. Capability inventory: Local file access, Python script execution, and LaTeX compilation. Sanitization: None described for the input text. While this surface exists, it is necessary for the skill's utility and no malicious patterns were identified.

cheatsheet