A humble library for finding related posts and content. Uses tf-idf under the hood. Primarily aimed at static site generators.

中文文本概要提取库（TextRank + 位置加权 + TF-IDF + MMR）

基于tf-idf-search代码改进, tf-idf-search不支持中文分词

Word statistics package (count, TF-IDF, etc.)

A package to extract important keywords from a document using TF-IDF technique

Search for terms in array of documents using TF-IDF

Knowledge base with BM25/TF-IDF search for GuideKit SDK

Generate BM25 and TF-IDF sparse vectors in JavaScript

Token-efficient memory, code indexing, and validation for Claude Code agents — SQLite + FTS5, TF-IDF + Qdrant retrieval, AST skeleton pruning, diff-aware context, Logic Guardian drift detection

Keyword extraction using TF-IDF and RAKE with synonym ring support (WASM)

inverted-index for level with pagination, sift3/cosine distance and tf-idf ranking

Framework-free resume / job-description scoring engine. Hybrid 60% semantic + 40% TF-IDF, runs locally via Transformers.js. Powers the ResuRank desktop app and the resurank-mcp MCP server.

Argo — 47-engine unified search MCP server with TF-IDF routing, RRF fusion, and Bocha reranker. 6 tools: search, research, evidence, clarify, crawl, extract.

Agentic OS MCP Server — 150+ tools with on-demand loading (26 core), Streamable HTTP transport, background task engine, browser automation, TF-IDF semantic memory, multi-agent orchestration. Zero API keys.

Plugin to add TF-IDF scores to a Lyra index

Japanese text vectorization using TF-IDF + LSA with n-gram tokenization and custom dictionary support

Covectric is a simple vector based search engine using cosine similarity and tf-idf methods for finding text similarity.

Node package for TF-IDF which stands for Transform Frequency Inverse Document Frequency

A lightweight Node.js full-text search library with support for multi-field indexing and TF-IDF ranking.

Markdown compression + context extraction for LLM apps. Rule-based engine (up to 35% savings) + TF-IDF context extractor (up to 95% reduction on targeted queries).