openbotx
An open-source platform for orchestrating AI agents — secure, simple, and built for everyone. Multi-agent, real-time task board, web control panel, skills system, browser automation, multi-provider, scheduler,...
Uploads a video to YouTube using the official YouTube Data API v3 and OAuth 2.0. Use this skill when the user asks to upload a video to YouTube. It supports titles, descriptions, privacy settings (public, private, unlisted), and large file chunking. Requires a Google Cloud 'client_secret.json' file.
Follow the repository instructions for this skill and your agent. Review the files and dependencies before adding it to your workflow.
SkillHub CLI
skillhub install brorlandi/youtube-uploadInspect the source and compatibility requirements, then follow its installation steps.
Give your agent the source link and ask it to inspect instructions and dependencies before explaining the installation steps.
This skill allows you to securely upload videos to YouTube via the official API, bypassing the need for fragile browser automation.
pip3 install google-api-python-client google-auth-oauthlib google-auth-httplib2
client_secret.json file is required. The user must generate an OAuth 2.0 Client ID (Desktop App) from the Google Cloud Console with the YouTube Data API v3 enabled.Use the provided Python script to upload the video:
python3 scripts/upload.py \
--file "/path/to/video.mp4" \
--title "My Video Title" \
--description "My Video Description" \
--privacy "unlisted" \
--secrets "/path/to/client_secret.json"
On the very first run, the script will output a URL or open a browser window for the user to authenticate and grant permission to their YouTube account. Instruct the user to complete the login flow. Once approved, a token.pickle file is generated locally, and subsequent uploads will run silently.
token.pickle and re-run to trigger the auth flow again.These signals come from indexed source analysis and may lag behind changes. Check the current source files.
A specialized skill for OpenClaw that enables AI agents to securely upload videos to YouTube using the official YouTube Data API v3 and OAuth 2.0.
This skill bypasses the need for fragile browser automation (like Playwright/Puppeteer) by utilizing a robust Python script under the hood, supporting large file uploads (chunking), titles, descriptions, and privacy settings.
google-api-python-client with resumable uploads to handle large video files without crashing..skill, allowing agents like Claude Code or Gemini to seamlessly trigger YouTube uploads.--title, --description, and --privacy (public/private/unlisted) arguments.Google Cloud Credentials:
You must generate an OAuth 2.0 Client ID (Desktop App) from the Google Cloud Console with the YouTube Data API v3 enabled.
Save the downloaded JSON file as client_secret.json in the root of the skill folder.
Python Dependencies: The host machine needs Python 3 and the following packages installed:
pip3 install google-api-python-client google-auth-oauthlib google-auth-httplib2
To install this skill locally in your OpenClaw environment:
openclaw skills install <path_to_skill_folder_or_.skill_file>
On the very first run, the script requires user interaction. The agent will execute the script, which will generate an OAuth URL. The user must click the URL, authenticate with their Google account, and grant YouTube upload permissions.
Once approved, a token.pickle file is generated locally, and all subsequent uploads will run silently and automatically.
Agents will internally call the script like this:
python3 scripts/upload.py \
--file "/path/to/video.mp4" \
--title "My Awesome...
# YouTube Upload Skill This skill allows you to securely upload videos to YouTube via the official API, bypassing the need for fragile browser automation. ## Prerequisites 1. The Google API Python Client and OAuth libraries must be installed: ```bash pip3 install google-api-python-client google-auth-oauthlib google-auth-httplib2 ``` 2. A `client_secret.json` file is required. The user must generate an OAuth 2.0 Client ID (Desktop App) from the Google Cloud Console with the YouTube Data API v3 enabled. ## Usage Use the provided Python script to upload the video: ```bash python3 scripts/upload.py \ --file "/path/to/video.mp4" \ --title "My Video Title" \ --description "My Video Description" \ --privacy "unlisted" \ --secrets "/path/to/client_secret.json" ``` ### First Run (Authentication) On the very first run, the script will output a URL or open a browser window for the user to authenticate and grant permission to their YouTube account. Instruct the user to complete the login flow. Once approved, a `token.pickle` file is generated locally, and subsequent uploads will run silently. ## Troubleshooting - **Token Expired / Revoked:** If the token becomes invalid, delete `token.pickle` and re-run to trigger the auth flow again. - **Quota Exceeded:** The YouTube API has a daily upload quota. If this is hit, the user must wait until the quota resets.
# OpenClaw YouTube Upload Skill A specialized skill for OpenClaw that enables AI agents to securely upload videos to YouTube using the official YouTube Data API v3 and OAuth 2.0. This skill bypasses the need for fragile browser automation (like Playwright/Puppeteer) by utilizing a robust Python script under the hood, supporting large file uploads (chunking), titles, descriptions, and privacy settings. ## Features - **Reliable Uploads:** Uses `google-api-python-client` with resumable uploads to handle large video files without crashing. - **Agent Ready:** Packaged as an OpenClaw `.skill`, allowing agents like Claude Code or Gemini to seamlessly trigger YouTube uploads. - **Customizable:** Agents can pass `--title`, `--description`, and `--privacy` (public/private/unlisted) arguments. ## Prerequisites 1. **Google Cloud Credentials:** You must generate an OAuth 2.0 Client ID (Desktop App) from the Google Cloud Console with the **YouTube Data API v3** enabled. Save the downloaded JSON file as `client_secret.json` in the root of the skill folder. 2. **Python Dependencies:** The host machine needs Python 3 and the following packages installed: ```bash pip3 install google-api-python-client google-auth-oauthlib google-auth-httplib2 ``` ## Installation for OpenClaw To install this skill locally in your OpenClaw environment: ```bash openclaw skills install <path_to_skill_folder_or_.skill_file> ``` ## First-time Authentication On the very first run, the script requires user interaction. The agent will execute the script, which will generate an OAuth URL. The user must click the URL, authenticate with their Google account, and grant YouTube upload permissions. Once approved, a `token.pickle` file is generated locally, and all subsequent uploads will run silently and automatically. ## Usage (CLI / Internal) Agents will internally call the script like this: ```bash python3 scripts/upload.py \ --file "/path/to/video.mp4" \ --title "My Awesome...
An open-source platform for orchestrating AI agents — secure, simple, and built for everyone. Multi-agent, real-time task board, web control panel, skills system, browser automation, multi-provider, scheduler,...
OpenClaw skill for 知识星球 / ZSXQ digests with token-first auth and browser recovery.
🌐 OpenClaw skill for Browser-Use — AI-powered browser automation for complex multi-step workflows (login, form filling, scraping, posting)
OpenClaw skill to crosspost X/Twitter posts to Reddit via browser automation