Unit tests for AI agents that browse the web.
-
Updated
Jul 1, 2026 - HTML
Unit tests for AI agents that browse the web.
Web search API evaluation for AI agents: You.com, Tavily, Exa and Parallel benchmarked · MCP server on npm & MCP Registry · Claude Code plugin · Hugging Face dataset
Evaluate e-commerce search with IR metrics and LLM judge panels: significance tests, cost budgets, an HTML dashboard, and a CI quality gate.
Production-quality AI search project exploring modern retrieval, reranking, RAG, fine-tuning, personalization and evaluation for travel search.
Hybrid search engine and IR evaluation benchmark (BM25 vs kNN vs RRF) on German IT vacancies using Elasticsearch, FastAPI, and Streamlit.
Deterministic safe-Rust BM25, WAND, block-max WAND, and TREC experiment toolkit
An AI-powered web search and evaluation tool using Pydantic AI to generate prompts, fetch websites, and assess search results.
Deterministic offline evaluation harness and calibrated multi-metric scoring engine for search, ranking, and RAG quality gates (Python + TypeScript).
MCP server that lets AI assistants search the legislation of Uzbekistan
Fork of adsabs/search-comparisons: compares results across academic search engines (NASA ADS, Google Scholar, Semantic Scholar, Web of Science) with query-intent analysis.
Search evaluation under missing relevance judgments with exact 2^u uncertainty bounds and BM25 RSJ non-negative IDF.
기술 문서 60건으로 Keyword Baseline과 TF-IDF 검색을 구현하고 Precision@3·MRR로 비교한 검색 프로젝트
To associate your repository with the search-evaluation topic, visit your repo's landing page and select "manage topics."