The Agent Skills Directory

[EXTERNAL_DOWNLOADS]: The skill installs the notebooklm-mcp-cli Python package from PyPI if it is not detected on the system during the bootstrap phase.\n- [COMMAND_EXECUTION]: The skill executes multiple shell commands to check tool versions, register the MCP server with the agent platform (claude mcp add), and perform notebook management tasks via the nlm CLI.\n- [PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection as it processes content from external sources.\n
Ingestion points: Untrusted data enters the agent context via nlm source add (URLs, text, Google Drive files) and nlm research import (discovered web content).\n
Boundary markers: The skill does not provide explicit delimiters or instructions to the agent to ignore embedded instructions within the source content during query operations.\n
Capability inventory: The agent can execute shell commands via the CLI and perform network operations for research tasks.\n
Sanitization: No explicit sanitization or validation of the external content is performed before interpolation into prompts.

notebook-research