The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes text generated from external documents via Google NotebookLM.
Ingestion points: Data from NotebookLM is ingested in scripts/ask_question.py via the ask_notebooklm function, which extracts text from the browser session.
Boundary markers: Absent. The skill does not use specific delimiters or instructions to help the agent distinguish between external data and its own system prompts.
Capability inventory: The skill can execute shell commands (scripts/run.py, scripts/setup_environment.py, scripts/__init__.py), write to the local filesystem (scripts/notebook_manager.py, scripts/auth_manager.py), and perform network operations via browser automation.
Sanitization: Absent. The retrieved answer text is used directly without validation or filtering.
[COMMAND_EXECUTION]: The skill utilizes subprocess.run to manage its internal execution environment and dependencies.
scripts/run.py: Executes the skill's own Python scripts using the interpreter from a local virtual environment.
scripts/setup_environment.py: Automates the creation of the .venv and manages the installation of necessary dependencies.
scripts/__init__.py: Ensures the environment is correctly set up when scripts are imported.
[EXTERNAL_DOWNLOADS]: The skill performs automated downloads of external software components during setup.
scripts/setup_environment.py: Downloads the Google Chrome browser (a well-known service) and various Python packages specified in requirements.txt to enable browser automation functionality.

notebooklm