Show HN: TinySearch – token-efficient web research for local AI agents

Category: infrastructure

Tags: web-research, ai-agents, mcp-server

Score: 7.0/10 (Innovation: 6, Technical: 7, Documentation: 8, Utility: 7)

TinySearch is a self-hosted web research tool for local AI agents, enabling search, crawling, reranking, and source-grounded prompt generation via MCP. Its clever integration of ONNX embeddings, SearXNG, and a structured pipeline for token-efficient context extraction makes it a practical enhancement for agentic workflows.

Target audience: backend devs, ai engineers, devops

Repository: https://github.com/MarcellM01/TinySearch · Python · MIT · 105 stars

View on Hacker News