The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes local Python scripts and shell wrappers to interface with the NotebookLM system and internal memory managers.
Evidence: SKILL.md and scripts/ask_design.sh invoke python skills/notebooklm/scripts/run.py and python3 execution/memory_manager.py.
[EXTERNAL_DOWNLOADS]: The skill is designed to interact with Google's NotebookLM service to retrieve design documentation.
Evidence: references/usage_patterns.md references the notebooklm.google.com domain for RAG operations.
[PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes retrieved data from an external RAG source (NotebookLM).
Ingestion points: User-provided questions and external content retrieved via ask_question.py in SKILL.md.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the skill body.
Capability inventory: Command execution capabilities are present via Python and Bash scripts mentioned in SKILL.md.
Sanitization: No explicit sanitization or filtering of the RAG-sourced content is described in the prompt instructions.

claude-code-design