Custom agents

A CLI that takes natural language and converts it to bash commands using a local code generation model or a curated prompt-based rule engine.

Pure-Python, zero-dependency RAG memory engine for conversational AI. Retrieves semantically relevant messages from conversation history using two-phase retrieval, TF-IDF + concept-overlap scoring, narrative element extraction, and bidirectional typo correction. No embeddings or vector DB required.

httpie for embeddings. Embed text, images, audio, video, and PDFs from the command line.

[Experimental] Open-source RAG System for diverse document types

A runtime governance, security and execution control layer for Retrieval-Augmented Generation (RAG) systems

A genral RAG Search chatbot, with SoTA RAG techniques such as HyDE, Hybrid retrieval with BM25 + RRF and Cross encoder reranking. Evaluated on the BEIR scifact dataset and compared all the different pipelines i tried along the way

Published PyPI package for ArXiv embedding benchmarks, retrieval evaluation, and scientific RAG experiments.

Open-source machine-readable standard for pharmaceutical SOPs - deterministic, auditable, GxP-aligned

implemented vector similarity algorithms to understand their inner workings, used local embeddding models

API-difycall 是一个轻量级的 Python 项目，用于简化与 Dify API 的交互。Dify 是一个强大的 AI 应用开发平台，本库旨在让开发者能够更便捷地集成和使用 Dify 的各项功能。

SOC Analyst Automation using a RAG model integrates a knowledge retrieval system with generative AI to automate SOC Level-1 tasks. It processes server logs, retrieves relevant security insights, and generates accurate responses, enhancing incident analysis, reducing response times, and improving efficiency in handling cybersecurity threats through

An Intelligent SOC Ecosystem that integrates SIEM, SOAR, and SmartXDR to enhance monitoring, detection, and response to cyber attacks. SmartXDR combines OpenXDR tools (Wazuh, Suricata, Zeek, pfSense,…) with AI/ML to reduce false positives, prioritize critical threats, and accelerate incident response, ensuring a more adaptive and effective SOC.

A lightweight, local Retrieval-Augmented Generation (RAG) system for domain-specific Q&A over financial documents. Uses pdfplumber for PDF parsing, sentence-transformers for dense retrieval, and optionally connects to local LLMs (e.g., Ollama + Mistral). Runs on FastAPI with a custom frontend.

Prompt-safety guards as a Python library. TF-IDF, RAG, LLM as a judge pipeline

💬 Engage in continuous chat while dynamically building a JSONL playbook, enhancing insights and strategies with each interaction.

Simple tool for RAG (Retrieval Augmented Generation) applications to be run over locally stored files

This RAG Streamlit app lets users chat with PDF documents using Gemini and Google's generative AI. Upload PDFs, process text, and get intelligent answers to your questions.

This Graph RAG Application is a web-based tool that allows users to ask questions about the Mission Impossible film franchise and receive detailed, contextually relevant answers. By combining retrieval-based methods with generative AI, the application ensures that responses are both accurate and engaging.

Code for the presentation.

A Chrome extension that performs real-time Retrieval-Augmented Generation (RAG) on any webpage. It extracts page content, processes it into chunks, and uses a Flask-based backend to perform vector searches and generate context-aware answers using LLM

Platform for RAG with NASA data for managing data on exoplanets for Nasa Space App Challenge 2025.

The AI-Based Pre-Screening Assistant Bot is an AI-powered technical interview assistant that evaluates candidates using adaptive questioning, real-time feedback, and automated report generation. Built with Streamlit and RAGPipeline, it dynamically adjusts question complexity, tracks candidate performance, and emails interview summaries

EarningsAI Demo is a powerful tool that combines audio transcription, document processing, and AI-powered analysis to help users extract insights from earnings calls and financial documents. Built with Fireworks AI and MongoDB, it provides both a command-line interface and a web application for processing and querying financial data.

Advanced RAG architecture built using Transformers and TensorFlow to enabled credibility score checker for social media posts. Recent articles are web-scrapped from google-news-api using scrapy and stored in MongoDB, their titles are embedded in ChromaDB. This is accessed by LLM using Mistral API key.

RAG agent with SGR, smart indexing and reranking, metadata enrichment, parent document extraction

BigQuery Semantic Search Orchestrator

AI assistant that uses RAG architecture and responds instantly with support from university regulations.

The document normalization engine RAG has always needed. Parse any document, understand its structure, build RAG that actually works.

RAG-optimized documentation and data sources for AI-native development on the Flow network. Ground LLMs in accurate Flow and Cadence context.

A small reranker service using mixedbread.ai reranker model

A project to show good CLI practices with a fully fledged RAG system.

A meta-app, built with Streamlit—supercharged with Snowflake AI—to help Streamlit evolve, one issue at a time.

RAG code sorting search, RAG knowledge organization

A comprehensive collection of RAG (Retrieval Augmented Generation) implementations 📚✨, from foundational concepts to advanced agentic 🤖 and knowledge graph 🌐 RAGs

Baseline replication for paper “EMERGE: Enhancing Multimodal Electronic Health Records Predictive Modeling with Retrieval-Augmented Generation.”

Production-ready Snowflake RAG system with type-specific chunking

There is documents searcher project based on Rust and Opensearch technologies.

Implementation of Retrieval Augmented Generation RAG-based LLM from html web pages

Repo for DermAssist: Your AI Assitant for Skin Problems. Powered by a vision model and a reliable RAG system.

A comprehensive evaluation toolkit for assessing Retrieval-Augmented Generation (RAG) outputs using linguistic, semantic, and fairness metrics

A local RAG pipeline for the Hollow Knight wiki - chat with wiki knowledge using Ollama and Supabase.

Retrieval Pivot Attacks in Hybrid RAG (Vector → Graph)

Raku package for doing LLM Retrieval Augmented Generation (RAG).

RAG - Ops: RAG with Operations. We help you to deploy RAG with changing in datasources.

A system that combines Retrieval-Augmented Generation (RAG), the Claude Sonet 3.5 LLM, and the Pathway framework to analyze financial reports and tables. It ingests data from Google Drive, processes both structured and unstructured formats in real time, and presents insights via a Streamlit-powered interface.

AI based conversational language tutor

LLM Chatbot-RAG by OpenVINO. The chatbot can read a PDF file and answer to the related questions.

demo in blog how ai help understand omniclass

RAG Evaluation using RAGAS

SwiftUI app that uses local RAG over git diffs to help generate commit messages with Ollama.

Llama_RAG_System is a local Retrieval-Augmented Generation (RAG) system that leverages the LLaMA model to provide intelligent answers to user queries by processing uploaded PDFs and fetching relevant web information while ensuring privacy.

Orion is an open-source AI native self-learning platform. Write your content, query it using a agents, access your knowledge base.

A powerful platform designed to simplify the creation and management of Data Contracts, bridging systems for seamless data ingestion

Data Pipeline with validation + AI Assistant

A chatbot that provides insights into the luxury fashion industry. It uses Cohere API and ChromaDB for efficient document retrieval and dynamic conversation.

Retrieval Augmented Generation (RAG)

Fashion authentication agent: LangGraph + GPT-4o Vision + RAG. Paste an eBay link, get a confidence score.

C# SDK for the Tavily API -- AI-powered search and extraction

AI onboarding copilot with RAG, escalation levels, and MLflow LLMOps. FastAPI · Streamlit · LanceDB.

A powerful knowledge management system that forges wisdom from experiences, insights, and best practices. Built with Qdrant vector database for efficient knowledge storage and retrieval.

Turn any LLM into a self-extending knowledge agent powered by a graph-structured memory - complete with PDF-to-graph ingestion, budget-aware optimisation, and dual-engine orchestration.

3D solar system explorer

Correct complete RAG -- built for Highway Workflow Engine

Native macOS personal knowledge base with AI — LightRAG knowledge graph, RAG chat, SpriteKit graph viewer, MCP bridge, 6 LLM providers. Privacy-first: your data stays local.

NLP + RAG technique for requirements engineering.

Moved to github.com/vanedb/vanedb (cpp/) — archived

Using LLMs to generate custom UI elements. Currently using Mistral-7b.

🎲 RuleLawyer – An AI-powered assistant that helps you look up and understand board game rules using Retrieval-Augmented Generation (RAG). Upload your rulebooks and ask it anything!

Mindsage is an all offline journalling app

Python asynchronous SDK for managing RAGFlow APIs

Curso completo de AI Engineer en español: 6 módulos / 16+ sesiones, TypeScript-first multi-provider, proyecto integrador real. MIT + CC BY 4.0.

This demo demonstrates the integration of a local LLama3 model for handling JSON responses within the Spring AI framework, positioning the model as a 'business logic' executor

IMDb IR System Project - MIR Course - Spring 2024

ACL Findings 2025 | RaPID: retrieval-augmented long text generation with planning and information discovery.

HealthMate is an AI-powered voice agent that helps users get clear, reliable answers to health-related questions in real time. It combines FastAPI, AssemblyAI, and React with Retrieval-Augmented Generation (RAG) and VectorDB Cloud (Native) for fast and relevant responses.

[λ] Eigen - the platform to search your documents without limits

A technical white paper on how LLMs handle memory, why context windows alone are not enough, and what production engineers need to know about memory architectures, security risks, and the road to in-weights personalisation.

A 100% offline, multi-agent AI swarm powered by Ollama. Features 5 specialized agents, hybrid RAG memory, and ChatDev integration. Protected by a strict 4-tier safety execution protocol to safely unleash uncensored Protocol 666 and offensive security Hat Mode algorithms.

Reproducible UFO/UAP archive extraction: OCR, cleaned text, RAG chunks, and knowledge graph for the War.gov release.

RAG的最小实现

Automates discovering and categorizing new Neovim plugins with RAG, creating PRs to yutkat/my-neovim-pluginlist.

Practical sessions for the “Data Science for Economics” at the CEMFI Summer School 2025.

Insightic is a comprehensive digital forensics platform with MongoDB-based case management, real-time map visualization, and AI-powered analysis capabilities.

RAG Model used to further improve GPT models to provide accurate, up to date, information to users

RAG server in ollama and golang

Python client for Eclipse RDF4J — interact with RDF4J repositories, execute SPARQL queries, and manage RDF data seamlessly in Python.

Your terminal forgets everything. engram remembers. Ask it questions about past errors, deployments, or any command you've ever run.

Fault-tolerant multi-agent AI architecture: hierarchical agent coordination, edge-cloud hybrid inference, and adversarial chaos testing for systems that can't afford to fail.

An AI/LLM/RAG/Agent glossary for everyone interested in AI | AI 专业名词百科，用不吓人的语言帮你看懂 AI 世界

RAG Backend for Aleph Alpha LLMs.

基于书生浦语的农业助手

利用向量数据库和GPT搭建的本地知识库问答应用

Rust CLI tool to interact with Ollama with Surreal DB Vector Database

Run. Sip. Repeat. This isn’t just bar hopping—it’s bar running. Let’s make your night legendary.

Stop writing resumes from scratch. EvidenceCV turns your real work into an ATS-optimized resume using RAG. 🥉 3rd Place - eHacks Hackathon 2026.

OutcomeOps: Context Engineering for AI-Assisted Development

Token-efficient, layered context delivery for AI agents. Four memory tiers (Identity, Session, Experience, Archive) — context is always available, just collapsed by default.

Native Rust PKM for markdown notes with local AI. Iced desktop GUI, terminal CLI, and TUI — semantic search and cited RAG chat via Ollama. Fully local.

The first open-source MCP server for jPOS and ISO 8583. Give AI agents deterministic access to payment protocol expertise.