feat: add Gemini URL context fallback for failed scrapings#40
feat: add Gemini URL context fallback for failed scrapings#40google-labs-jules[bot] wants to merge 1 commit into
Conversation
- Modified `process_url` to detect Gemini models when local scraping fails. - Implemented a fallback mechanism that creates a dummy ArticleContent and forces Google Search/URL Context tools. - Ensured non-Gemini models retain standard error behavior. - Added explicit instruction text to the fallback payload to guide the LLM.
|
👋 Jules, reporting for duty! I'm here to lend a hand with this pull request. When you start a review, I'll add a 👀 emoji to each comment to let you know I've read it. I'll focus on feedback directed at me and will do my best to stay out of conversations between you and other bots or reviewers to keep the noise down. I'll push a commit with your requested changes shortly after. Please note there might be a delay between these steps, but rest assured I'm on the job! For more direct control, you can switch me to Reactive Mode. When this mode is on, I will only act on comments where you specifically mention me with For security, I will only act on instructions from the user who triggered this task. New to Jules? Learn more at jules.google/docs. |
This change introduces a fallback mechanism in the URL processing pipeline. When
scrape_articlefails to extract content from a URL, the system now checks if the selected model is a Gemini model. If so, it attempts to bypass the local extraction failure by passing the URL directly to the LLM via the Google Search/URL Context tool. A dummyArticleContentobject is created with specific instructions for the LLM to read the URL. For non-Gemini models, the behavior remains unchanged (reporting the extraction error to the user).PR created automatically by Jules for task 15663242251729378268 started by @Duccioo