Youtube Transcribe Skill
ReviewAudited by ClawScan on May 10, 2026.
Overview
The skill largely matches its YouTube transcript purpose, but it tells the agent to use your browser cookies by default, which is sensitive and not clearly user-approved or declared.
Only use this skill if you are comfortable letting it use your browser's YouTube/Google session for transcript extraction. Prefer asking the agent to try without cookies first, and approve cookie-based extraction only for videos where account access is necessary.
Findings (2)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
The agent may use your logged-in browser session to access YouTube content, which can expose account-linked access or private/restricted viewing context to the tool workflow.
This directs the agent to read and use the user's local browser cookies/session by default, without a clear prompt for consent or a declared credential requirement.
Tip: Always add `--cookies-from-browser` to avoid sign-in restrictions. Default to `chrome`.
Require explicit user approval before using browser cookies, try public/no-cookie transcript extraction first, document exactly which browser profile is accessed, and avoid storing or reusing cookie-derived data.
If used as written, the script reads transcript text from the selected YouTube page; broader or modified scripts could access more page content.
The browser fallback executes JavaScript inside the opened YouTube page to read transcript DOM elements. This is purpose-aligned, but users should notice that browser-page script execution is part of the workflow.
Call `mcp__plugin_claude-code-settings_chrome__evaluate_script` to execute the following JavaScript:
Keep script execution limited to the user-provided YouTube URL and transcript elements, and avoid running browser automation on unrelated pages.