Skip to content
YTScriptPro

Settings

Preferences are saved in this browser. Provider keys live on the server and are never shown here.

Appearance

Generation defaults

Used to pre-fill the generator and tools.

Used for word targets and duration estimates. Read a paragraph aloud for a minute and count the words to find yours.

Editor

Save changes automatically while you type. When off, press Save or Ctrl/⌘+S.

AI provider

Generation runs on this website's server using the provider its operator configured.

Configuration status

Checking…

Configure a provider to enable the test.

Choosing and configuring a provider

The provider is chosen on the server with environment variables (in .env.local or your host's settings), never in the browser — so keys can't leak to visitors. Restart the server after changing them.

  1. Create an API key in the Anthropic Console.
  2. Pick a model ID from Anthropic's models overview and set it as ANTHROPIC_MODEL. Faster, lower-cost models suit high-volume tools; larger models may write better long scripts.
  3. Add the variables below, restart the server, then use Test connection.
AI_PROVIDER=anthropic
ANTHROPIC_API_KEY=your-key-here
# A current model ID from Anthropic's API documentation
ANTHROPIC_MODEL=your-model-id

Costs and data

Hosted APIs bill the key owner per request, and your scripts are processed by the provider. On a public deployment, anyone who can reach the site can spend your quota — see the rate-limiting notes in the README.

Troubleshooting

“Not configured”
Check spelling of the variables, that AI_PROVIDER matches the variables you set, and restart the server.
Authentication errors
The key is wrong, revoked, or lacks access to the model. Generate a new key and update the environment.
Model not found
The model ID is misspelled or unavailable to your key. For local servers, make sure the model is installed and the base URL ends with the API prefix (often /v1).
Could not connect
The model server isn't running, is listening on a different port, or isn't reachable from where this app runs (see networking above).
Timeouts
Long scripts take time, especially on local hardware. Increase AI_REQUEST_TIMEOUT_MS or choose a shorter duration.
Output cut off
Raise AI_MAX_OUTPUT_TOKENS (within your model's limits) or generate shorter scripts.

Local data

Projects are stored only in this browser. They don't sync between devices or browsers, aren't encrypted by the app, and clearing site data removes them. Export a backup regularly to keep a copy.

Counting projects… stored in this browser.