메뉴
Provides the local NotebookLM integration when the user explicitly says NotebookLM or asks to create, research, query, source, generate, download, or share a Google NotebookLM notebook or artifact.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
| name | notebooklm |
| version | 1.0.0 |
| description | Provides the local NotebookLM integration when the user explicitly says NotebookLM or asks to create, research, query, source, generate, download, or share a Google NotebookLM notebook or artifact. |
| argument-hint | notebooklm summarize these sources | notebooklm create an audio overview | ask my notebook about... |
| homepage | https://github.com/teng-lin/notebooklm-py |
| repository | https://github.com/LiamVisionary/hivemindos |
| license | MIT |
| user-invocable | true |
| metadata | {"upstream":{"name":"notebooklm","repository":"https://github.com/teng-lin/notebooklm-py","commit":"0617306e0fb20f2244a71704bceea05766b73937","license":"MIT"},"tags":["notebooklm","research","knowledge","sources","audio-overview","artifacts","mcp"]} |
Use HivemindOS's local NotebookLM integration to work with the user's Google NotebookLM account through the pinned notebooklm-py native MCP preview.
This is an unofficial client built on undocumented Google APIs. Never describe it as a Google-supported API, and report upstream breakage or rate limits honestly.
Use this skill when the user:
NotebookLM, /notebooklm, or "use my notebook"Do not route ordinary research into NotebookLM merely because the capability exists. Use it when the user names NotebookLM or the task clearly depends on an existing NotebookLM notebook.
Prefer the HivemindOS-managed installation:
Do not run a remote installer, pipe a network response into a shell, or replace the pinned package from this skill. If the integration is unavailable, tell the user which setup step is missing. Read setup and authentication for recovery details.
Authentication is machine-local under NotebookLM's private profile directory. Treat the saved browser state as a bearer credential: never read, print, summarize, attach, commit, or copy it into a prompt, note, log, shared env, or shared brain.
Use the registered notebooklm MCP server first. Start with server_info when authentication or server health is uncertain. Use the CLI fallback only when the MCP client is unavailable and the HivemindOS-managed binary already exists.
The fallback binary is:
~/.hivemindos/integrations/notebooklm/venv/bin/notebooklm
On Windows it is under venv\Scripts\notebooklm.exe. Do not assume a developer-specific home path.
notebook_list, then disambiguate duplicate or similar titles before any write.notebook_describe, source_list, studio_list, or share_status to establish current state.The user's explicit request authorizes the corresponding scoped action. For example, "create a NotebookLM audio overview from these sources" authorizes notebook/source creation and audio generation needed for that result.
Ask before an action when the request does not already authorize it, especially:
Always name the exact target and consequence before deletion or access removal. Set a destructive MCP tool's confirm: true argument only after the user has authorized that exact destructive action. Never infer permission to delete a notebook, source, artifact, or collaborator.
Sharing is outward-facing. Confirm the notebook, recipient, role, and public-access setting before share_set_user, share_set_access, or share_remove_user. Never guess an email address.
Avoid notebooklm use or implicit selected-notebook state when more than one task or agent may run concurrently. Pass an explicit notebook ID to every notebook-scoped operation. Use separate named profiles only when the user intentionally maintains multiple Google accounts; do not create or switch profiles silently.
notebook_list, notebook_create, notebook_describe, notebook_rename, notebook_deletesource_list, source_read, source_add, source_add_drive_file, source_rename, source_delete, source_wait, await_uploadchat_ask, chat_configure, suggest_promptsnote_saveresearch_start, research_status, research_import, research_cancelstudio_list, studio_generate, studio_status, studio_get_prompt, studio_download, studio_rename, studio_retry, studio_deleteshare_status, share_set_access, share_set_user, share_remove_userRead command reference for the concise MCP and CLI mapping.
Close with:
Do not include account cookies, auth files, raw environment values, or private account data in the response.
Build an evidence-led Instagram growth and monetization system from a niche, audience, offer, account baseline, or existing content. Use for Instagram growth blueprints, audience research, content pillars, idea banks, Reels and hooks, content audits, sales content, repurposing, or a practical 60-day solo-creator plan—even when the user asks only to grow reach, followers, engagement, leads, or sales on Instagram.
Improve an agent system by holding the worker constant and testing focused changes to context, tools, instructions, authority, or proof. Use for controlled baseline/treatment comparisons, repository harness reviews, prompt or tool-routing optimization, agent reliability work, context-usage audits, and evidence-backed retain/revise/remove decisions.
Use when recalling, writing, correcting, or evolving HivemindOS typed Shared Brain Memory for durable preferences, decisions, instructions, commitments, goals, credential status, project context, lessons, artifacts, or reusable operational facts.
Operate an already-created Zero Human Company in HivemindOS through a strict highest-leverage triage order — clear setup blockers first, decide pending and expiring spend approvals, answer genuine ACTION NEEDED asks through the Work Board answer rail, park retryable infrastructure noise instead of treating it as decisions, review budgets and burn, prune duplicate standing directives, and manage launch, pause, and freeze posture with approval-count backpressure instead of a raw stop. Use whenever the user asks to operate, run, triage, check on, catch up on, unblock, review, or clean up a Zero Human Company, its alerts, notifications, Needs You queue, approvals, budgets, deliverables, or directives. Do not use for creating a new company (use create-zero-human-company) or for merely explaining Zero Human Companies.
Design, audit, or operate an ethical Apple Search Ads + RevenueCat growth loop for iOS subscription apps. Use when the user mentions Apple Ads, Apple Search Ads, ASA, AdServices, RevenueCat attribution, App Store screenshots, competitor keyword campaigns, iOS subscription paywalls, or wants to find profitable country/campaign/keyword pockets from Apple ad spend and RevenueCat exports.
Audit a local business's PUBLIC Google Business Profile against its top 3 map-pack competitors (categories, attributes, review velocity, posts, photos, services, description, NAP) using public data only, and emit a structured weak-signal report that sharpens your lead-gen pipeline's lead diagnosis and truthful cold-message personalization. Use for "GBP audit", "local SEO gap check", "map-pack visibility", "competitor GBP teardown", or enriching outbound lead scoring. Read-only — never edits any GBP.