The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses subprocess.run to manage its environment and execute local scripts. Specifically, scripts/run.py and scripts/setup_environment.py invoke the Python interpreter and shell commands to set up a virtual environment and run automation tasks.- [EXTERNAL_DOWNLOADS]: The skill triggers the download and installation of external dependencies. It fetches Python packages from PyPI and the Google Chrome browser through the patchright library during its first-run setup.- [PROMPT_INJECTION]: The skill implements a follow-up mechanism by appending specific instructions to the responses fetched from NotebookLM. This 'FOLLOW_UP_REMINDER' in scripts/ask_question.py directs the agent to evaluate the completeness of the answer and potentially perform additional queries, which influences the agent's decision-making flow.- [PROMPT_INJECTION]: The skill processes unstructured text from Google NotebookLM, creating an indirect prompt injection surface.
Ingestion points: Text is extracted from the browser in scripts/ask_question.py (lines 104-142).
Boundary markers: Absent. The content is returned to the agent without specific delimiters or warnings to ignore embedded instructions.
Capability inventory: The skill can execute local Python scripts and shell commands via subprocess calls in scripts/run.py.
Sanitization: Absent. The content is retrieved and returned in its raw text form.

notebooklm