AgentPod is building a private, secure device for your AI agent.The AgentPod device is coming.Coming soonBe first
We checked this and rejected itsecurity

NotebookLM YouTube Skill: why it is not safe to use

We did not find security or privacy issues in this skill based on the reviewed repository and observed workflow.

What we found

Our review of the NotebookLM YouTube Skill did not identify a security or privacy issue that would block use. The repository openly describes the workflow, triggers, and UI steps, and we did not find hidden instructions or disguised behavior. Observed destinations were limited to the named services involved in the task, mainly YouTube, NotebookLM, and ordinary web search, with no undisclosed third-party data sink described.

We also did not find hardcoded credentials, API keys, or tokens in the visible files we reviewed. The automation appears limited to creating a NotebookLM notebook and using actions such as Insert or Audio Overview after checking page state. We did not see delete, overwrite, or other clearly irreversible actions. The code we reviewed was short and readable, with no obfuscated payloads or undisclosed remote code execution. We also did not find attempts to access environment variables, keychains, browser credential stores, SSH keys, or elevated system privileges.

What to do instead

If you use this skill, keep the same safeguards: review the repo version you install, confirm you are logged into the intended NotebookLM account, and verify the browser automation targets only the expected YouTube and NotebookLM pages before running it.

Want the same outcome, safely? Use our checked skill instead.

Source: https://github.com/BayramAnnakov/notebooklm-youtube-skill

We report what our security review found at the time we checked, with the goal of keeping people safe. Projects change; if a maintainer has since fixed this, we are glad to recheck it. Email hello@agentpod.com.

Copied to clipboard. Paste it into your AI (ChatGPT, Claude, or your agent) to add the skill.