Agent-Corex – Intelligent Tool Selection - Intelligent tool selection for LLMs – 50-75% cost reduction

Building LLM agents with 100+ tools? Context bloat kills performance and costs. Agent-Corex intelligently selects only the relevant tools your model actually needs. ⚡50-75% fewer tokens → massive cost savings 🚀 3-5x faster inference → better user experience 🎯 95%+ accurate tool selection → production-ready Hybrid Ranking Engine: • Keyword matching (<1ms) + semantic embeddings (50-100ms) • Works with any MCP server Use cases: autonomous agents, multi-step reasoning, cost optimization.

Hi Product Hunt! 👋 We built Agent-Corex to solve a real problem we faced: context bloat in LLM agents. The Challenge: Most LLM-based systems include ALL available tools in the prompt. This causes: • Token bloat = higher API costs • Slower inference = worse UX • Model confusion = worse reasoning The Solution: Agent-Corex intelligently selects which tools to include using a hybrid approach: 1. Keyword ranking (<1ms) - Fast, zero deps 2. Semantic embeddings (50-100ms) - Accurate 3. Hybrid score - Best of both Real Impact: A team with 200 available tools saw: • 68% reduction in API costs • 4.6x faster inference • Same capability maintained Why We Built This: The problem scales linearly. At enterprise scale with millions of API calls monthly, tool selection alone can save $100K+ annually. What's Included: ✅ Open source (MIT license) ✅ 95%+ test coverage ✅ Production ready ✅ Works with Claude, GPT-4, Llama, any LLM ✅ RESTful API + Python SDK ✅ Docker/Kubernetes support We're early stage (v1.0.1) and looking for: • Early adopters to share feedback • Contributors to improve algorithms • Users for case studies Try it: pip install agent-corex GitHub: https://github.com/ankitpro/agen... Docs: https://ankitpro.github.io/agent... Happy to answer any questions! 🚀

Agent-Corex – Intelligent Tool Selection - Intelligent tool selection for LLMs – 50-75% cost reduction

Replies