update available models, filter: [FILTER|text]<\/div>\n
Fetches a list of available models from OpenAI and updates the model menu.
\nReturns a comma-separated list of models (up to 50) that match the filter.<\/p>\n
How it works:<\/strong><\/p>\n\n- Queries the OpenAI API for all available models.<\/li>\n
- Filters out non-chat models (embeddings, audio, vision, moderation, etc.).<\/li>\n
- Optionally filters by a substring (e.g., “gpt-4” to see only GPT-4 models).<\/li>\n
- Returns the list as text and updates the model dropdown menu.<\/li>\n<\/ul>\n
Examples:<\/strong><\/p>\n\n- update available models, filter: [gpt-4|text]<\/span> \u2192 Lists all GPT-4 models<\/li>\n
- update available models, filter: [\u200c|text]<\/span> \u2192 Lists all available chat models<\/li>\n<\/ul>\n
Tip:<\/strong> Use this block at the start of your project to discover new models as OpenAI releases them!<\/p>\n <\/p>\n
\n\u2699\ufe0f Response Configuration<\/h3>\nset max tokens [MAX_TOKENS|text]<\/div>\n
Sets the maximum length of ChatGPT’s responses in tokens.
\nTokens are word fragments – roughly 1 token = 0.75 words.<\/p>\n
How it works:<\/strong><\/p>\n\n- Higher values = longer, more detailed answers.<\/li>\n
- Lower values = shorter, more concise answers.<\/li>\n
- Range: 1\u20132048 tokens (the extension keeps it within safe limits automatically).<\/li>\n
- Default: 360 tokens (~270 words) – keeps answers brief and kid-friendly.<\/li>\n
- GPT-5 Note:<\/strong> GPT-5 models use some tokens for internal \u201cthinking,\u201d which means they need a little extra max-token space to finish answers. This doesn\u2019t make their responses longer – it just supports their reasoning. The extension automatically increases their limit (\u00d72, minimum 240), but for very detailed answers you may still want to raise it further.<\/li>\n<\/ul>\n
Examples:<\/strong><\/p>\n\n- set max tokens [50|text]<\/span> \u2192 Very short answers (1\u20132 sentences)<\/li>\n
- set max tokens [200|text]<\/span> \u2192 Moderate-length answers (a few sentences)<\/li>\n
- set max tokens [500|text]<\/span> \u2192 Longer, detailed explanations<\/li>\n<\/ul>\n
Tip:<\/strong> Shorter answers are faster, cheaper, and easier for kids to read!<\/p>\n <\/p>\n
answers type: [TYPE]<\/div>\n
Controls the creativity and randomness of ChatGPT’s responses (temperature setting).
\nChoose from a menu of preset styles or enter a custom value (0\u20132).<\/p>\n
Options:<\/strong><\/p>\n\n- accurate<\/span> (0) – Focused, consistent, predictable answers<\/li>\n
- normal<\/span> (1) – Balanced creativity and accuracy (default)<\/li>\n
- creative<\/span> (2) – More varied, imaginative, surprising responses<\/li>\n<\/ul>\n
Examples:<\/strong><\/p>\n\n- answers type: [accurate]<\/span> \u2192 Good for math, facts, consistent responses<\/li>\n
- answers type: [creative]<\/span> \u2192 Good for storytelling, brainstorming, jokes<\/li>\n<\/ul>\n
<\/p>\n
set timeout [TIMEOUT|text]<\/div>\n
Sets the maximum time (in milliseconds) to wait for a response from OpenAI.
\nIf the request takes longer, it will timeout and return an error.
\nDefault: 30000 (30 seconds)<\/p>\n
Examples:<\/strong><\/p>\n\n- set timeout [60000|text]<\/span> \u2192 Wait up to 60 seconds<\/li>\n
- set timeout [15000|text]<\/span> \u2192 Wait up to 15 seconds (faster failure for slow connections)<\/li>\n<\/ul>\n
Tip:<\/strong> Longer timeouts are helpful for complex questions or slower internet connections.<\/p>\n <\/p>\n
\n\ud83d\uddd1\ufe0f Conversation Management<\/h3>\nclear message logs<\/div>\n
Clears the entire conversation history with ChatGPT.
\nUse this to start a fresh conversation without context from previous questions.<\/p>\n
When to use:<\/strong><\/p>\n\n- Starting a new topic or conversation.<\/li>\n
- When ChatGPT seems confused or is talking about old stuff.<\/li>\n
- To give your assistant a “fresh start” for testing or demos.<\/li>\n
- When you’ve had a long conversation (the extension remembers your last 10 questions & answers, then starts forgetting older ones).<\/li>\n<\/ul>\n
Note:<\/strong> Your assistant’s name, role, and personality stay the same – only the conversation memory is erased.<\/p>\n <\/p>\n
\n\ud83d\udd11 API Key Setup<\/h3>\nset api key<\/div>\n
Opens a prompt to enter or update your OpenAI API key.
\nYour key is stored in the browser’s session storage (cleared when you close the tab).<\/p>\n
How to use:<\/strong><\/p>\n\n- Click this block.<\/li>\n
- A text input prompt appears.<\/li>\n
- Paste your API key from OpenAI (starts with sk-proj-…<\/span>).<\/li>\n
- Click OK to save.<\/li>\n<\/ul>\n
Security Notes:<\/strong><\/p>\n\n- API keys are stored temporarily – only while your browser tab is open.<\/li>\n
- Keys are deleted when you close the browser tab.<\/li>\n
- Never share your API key publicly or in shared projects!<\/li>\n
- The extension automatically cleans up your API key – it removes invisible characters that sometimes get copied by accident.<\/li>\n
- Your key must use regular English letters and numbers only. If you paste something weird, you’ll get a friendly error message.<\/li>\n
- If you see an error about “invalid characters,” just copy the key again from OpenAI and paste it fresh.<\/li>\n<\/ul>\n
Tip:<\/strong> Generate separate API keys for different projects to track usage and costs separately.<\/p>\n <\/p>\n
\n\ud83c\udf93 Educational Uses<\/h2>\n\n- AI Literacy<\/strong> \u2013 Learn how AI models work, how prompts influence responses, and how to use AI responsibly.<\/li>\n
- Creative Learning<\/strong> \u2013 Build interactive tutors, storytellers, role-playing characters, and educational chatbots.<\/li>\n
- Language Skills<\/strong> \u2013 Practice real conversations in different languages with AI-driven characters.<\/li>\n
- API Understanding<\/strong> \u2013 Discover how Scratch connects to external services in a simple, visual way.<\/li>\n
- Physical Computing + AI:<\/strong>\n
\n- Natural-Language Robot Control<\/strong> \u2013 Command robots using phrases like \u201cdrive in a square\u201d or \u201cwave hello.\u201d<\/li>\n
- AI Decision Making<\/strong> \u2013 Combine sensors with ChatGPT so the robot decides what to do next.<\/li>\n
- Story-Acting Robots<\/strong> \u2013 ChatGPT writes scenes and the robot acts them out with motions or LEDs.<\/li>\n<\/ul>\n<\/li>\n<\/ul>\n