RAG · Search

It turns your document pile into a persistent, interlinked wiki so the LLM doesn't have to re-read everything every time you ask a question.

WeKnora exists to turn scattered enterprise documents into a queryable RAG, an autonomous reasoning agent, and a self-maintaining wiki.

Ruflo turns Claude Code from a solo assistant into a coordinated swarm with shared memory, plugins, and cross-machine federation.

This repo open-sources a full Chinese textbook on AI agent engineering—complete with Markdown source, compiled PDF, and runnable Python demos for every chapter.

Because stitching together LLM workflows, RAG, agents, and observability by hand is a full-time job.

Why rebuild the same RAG pipeline or agent loop from scratch when you can fork a working template instead?

It exists because AI context windows reset between conversations; Supermemory extracts facts, tracks changes, and recalls them later so agents don't start from scratch every time.

It centralizes model definitions so the same architecture works across PyTorch, JAX, vLLM, and llama.cpp without rewrites.

Persistent memory for coding agents that currently forget everything when the session ends.

Builds interactive knowledge graphs that teach you a codebase instead of just flexing its complexity.

LangBot is a Python framework that wires LLMs into every major chat app—WeChat, Discord, Slack, Telegram, Lark, DingTalk, QQ, and more—without making you rewrite adapters for each.

OpenDataLoader PDF exists to extract structured data from PDFs for AI pipelines while auto-tagging untagged documents for screen readers, all without proprietary dependencies.

JeecgBoot exists so Java teams can describe an ERP in plain language, generate the Spring Boot and Vue code, and then manually merge it to keep things flexible.

Mem0 gives AI agents a persistent memory layer that accumulates user context across sessions instead of overwriting it every time.

A CLI and library that turns prompt evaluation and red-teaming into repeatable, automatable engineering instead of vibe-based guesswork.

Rowboat exists because most AI assistants start every chat with amnesia, so it builds a persistent, local knowledge graph from your emails and meetings instead.

Official Jupyter notebooks demonstrating how to wire Claude into production tasks like RAG, SQL queries, and multimodal pipelines.

To wrap every LLM backend, RAG pipeline, and enterprise auth scheme into a single self-hosted interface.

A maintainer cataloged every Chinese NLP repo they touched into a single, obsessively categorized list so others wouldn’t have to hunt.

It exists so you can ask a database questions in English and let an agent write the SQL, run the Python, and generate the report without touching production unsupervised.