{"id":"claude-world-notebooklm-skill","name":"notebooklm-skill","af_score":55.0,"security_score":56.0,"reliability_score":30.0,"what_it_does":"notebooklm-skill bridges NotebookLM research with Claude-style content generation and distribution. It ingests URLs/PDFs/trending topics, creates and manages NotebookLM notebooks, runs deep research/QA with citations, generates multiple artifact types (text and downloadable assets like audio/video/slides/reports and some structured outputs), and can run as a CLI tool, a Claude Code Skill, or an MCP server for agent-driven workflows.","best_when":"You want an agent-friendly research-and-content pipeline that can be driven programmatically (MCP) or from the command line, with Google-browser login already acceptable.","avoid_when":"You need fine-grained authorization controls (scopes), strict error-code contracts, or strong operational guarantees like documented SLAs and robust retry/idempotency semantics.","last_evaluated":"2026-03-30T13:42:44.073894+00:00","has_mcp":true,"has_api":false,"auth_methods":["Browser-based Google login (via notebooklm-py session storage)","One-time CLI login that stores session state in ~/.notebooklm/storage_state.json"],"has_free_tier":false,"known_gotchas":["Session-based Google login is required; expired sessions may cause auth errors until re-login","Infographic is listed as excluded from downloadable types (possible mismatch with agent expectations around downloadable artifacts)","Audio format is M4A (not MP3), which can break downstream media tooling if assumed otherwise","No explicit rate-limit or retry/backoff strategy is described in the provided README"],"error_quality":0.0}