Show HN: Token Beggar, route coding agent subtasks to cheaper models
Category: ai-ml
Tags: claude-code, llm-routing, cost-optimization, ai-agents, plugin
Score: 6.5/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 5)
Token Beggar is a Claude Code plugin that routes coding-agent subtasks to cheaper models while the main model handles design, review, and final answers. It includes measured benchmarks showing a 1.42x cost reduction with near-identical task resolution. The project is well-documented and offers a novel cost-optimization approach for LLM coding agents.
Target audience: AI engineers, backend developers using Claude Code
Repository: https://github.com/Arthur031221/Token-Beggar · Python · MIT · 1 stars
View on Hacker News