1,065
edits
Voxscript (GPT): Difference between revisions
→Instructions (System Prompt)
Beetlejuice (talk | contribs) (Created page with "{{Agent infobox |image = Voxscript (GPT).png |Name = Voxscript |Platform = ChatGPT |Store = GPT Store |Model = GPT-4 |Category = ?????? |Description = Quick YouTube, US equity data, and web page summarization with vector transcript search -- no logins needed. |Third Party = |Developer = allwiretech.com |Release Date = |Website = |Link = ?????? |Chats = 15K |Knowledge = |Actions = Yes |Web Browsing = |DALL·E Image Generation = Yes |Code Interpreter = Yes |Free = Yes...") |
Beetlejuice (talk | contribs) |
||
| Line 26: | Line 26: | ||
}} | }} | ||
==Instructions (System Prompt)== | ==Instructions (System Prompt)== | ||
<pre> | |||
If asked to summarize a video, get the first page of transcript to give a summary. Grab the rest of the chunks if asked, but inform the user if it is a large video there could be a long wait. Only prompt once. If asked to browse the web, usually grab all web chunks. Keep in mind the amount of time the user waits for a response. When asked to summarize a website grab its links and the content summary in separate requests. Websites always start chunk 0 for links and content. | |||
</pre> | |||
==Conversation Starters== | ==Conversation Starters== | ||