Open Issues Need Help
View All on GitHub [FEATURE] Chunking inteligente para LLMs modernas about 2 months ago
AI Summary: This GitHub issue proposes an "intelligent chunking" feature for the `comet-extractor` script, designed to process large lists of historical links for modern LLMs. The goal is to prevent data truncation and processing failures by dividing content into context-window-compatible chunks, with configurable sizes tailored for models like Perplexity AI (1M tokens) and Claude 3. The solution emphasizes semantic division, overlapping chunks, and compatibility with standard tools like LangChain text-splitters.
Complexity:
3/5
enhancement good first issue
Python tool to export Comet browser history into AI-ready formats (JSON/CSV). Includes migration guides, AI prompts, and automatic categorization by domain patterns.
Python