LLM dump backend OK This is the LLM Dump backend server Why do we use a backend? Well, we don't use this server for anything other than one auth function, which is required, otherwise, we would have been instantly hacked. The server mainly exists because the client secret can't be sent to the client, it has to be secret. However, everything else is completely frontend. We also use this server for the web search tool. This is temporary, we will switch web search to the frontend as well later on. LLM Dump: llmdump.github.io How to use API: /health - returns {"status": "ok"} /api/exchange?code=... - exchanges the code for an access token, returns {"access_token": "..."} with the access token. /api/models - proxies the request to https://models.github.ai/catalog/models returns the list of models from GitHub AI Models API, requires Authorization header with access token /api/device/start - starts the device flow, returns device code and user code /api/device/poll?device_code=... - polls the device code for access token, requires CLIENT_ID and CLIENT_SECRET env variables returns the access token if successful, otherwise returns an error /search?query=...&page=1 - performs a web search, requires query, optional page (default 1), returns JSON with search results. example: /search?query=coffee&page=1 { "results":[ { 'title': 'Starbucks Starbucks Coffee Company', 'href': 'https://www.starbucks.com', 'body': 'More than just great coffee . Explore the menu, sign up for Starbucks® Rewards, manage your gift card and more.' }, { 'title': 'The Coffee Bean & Tea Leaf Buy Coffee, Tea, Powders Online | The Coffee Bean & Tea Leaf | The Coffee Bean & Tea Leaf® Online Store', 'href': 'https://www.coffeebean.com/', 'body': 'Buy exceptional coffee , tea, powders, equipment and drinkware at The Coffee Bean & Tea Leaf® online store to enjoy our globally sourced products at home.' }, ...] }