Feature: resolve web links in messages to embed text into prompt

[7flash]

Before:

After:

[mariodian]

Great idea! Although it only works with specific websites that are able to return a meaningful and short output.

Is there perhaps a tool that can summarize any web page of any size and you'd feed that to GPT?

[7flash]

Great idea! Although it only works with specific websites that are able to return a meaningful and short output.

Is there perhaps a tool that can summarize any web page of any size and you'd feed that to GPT?

There are such tools, and they can be used complementary with this feature by running local proxy server towards them. In fact, that's the main use case for me, - I am running caddy file server towards all my local codebase and then insert links pointing to files into prompt i.e. like "http://localhost:8080/home/repo/file.tsx"

[7flash]

@mariodian following example of Caddyfile I am using as a proxy server to my files

{
	admin 127.0.0.1:2023
}

:8080
root * /home/sevenflash/Code

header * {
	Access-Control-Allow-Origin "*"
	Access-Control-Allow-Methods "OPTIONS,HEAD,GET,POST,PUT,PATCH,DELETE"
	Access-Control-Allow-Headers "DNT,User-Agent,X-Requested-With,If-Modified-Since,Cache-Control,Content-Type,Range"
	Access-Control-Expose-Headers "Content-Length,Content-Range"
}

file_server {
	browse
	hide index.html
}

Then I can insert link "http://localhost:8080/galaxy-tauri-first/src/App.tsx" and it will be resolved into actual recent contents of the file.

[7flash]

can use this https://github.com/jina-ai/reader

Great idea! Although it only works with specific websites that are able to return a meaningful and short output.

Is there perhaps a tool that can summarize any web page of any size and you'd feed that to GPT?