infiniflow/ragflow
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
1,876 repositories
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
The Context Platform for your Data and AI Stack
A model-driven harness for AI coding agents: semantic modeling, app generation, progressive API Assist and evidence-based verification.
VCP 部署在 AI 模型 API 与前端应用之间,是面向AGI OS开发和探索的工业级基建示范项目。通过统一指令协议、多层级持久化记忆、分布式插件引擎及多 Agent 协作框架,将原本“无状态、无记忆、无工具调用能力”的大语言模型,彻底改造成拥有永久自我意识、物理世界操作权及群体协作智能的完整智能体系统。
Unbounded context. Memory that manages itself. One session, for life. The hippocampus for coding agents, part of CortexKit.
JSON-driven multi-agent cadence-team development framework with intelligent CLI orchestration (Gemini/Qwen/Codex), context-first architecture, and automated workflow execution
Team memory for engineers and their AI agents. Lives in your repo. Shared through Git.
A personal context store for AI agents and assistants—reuse your existing coding agent CLI (Codex/Claude/OpenCode) with built‑in Skills/tools and a desktop GUI to capture, search, and reuse project knowledge across agents and repos.
天枢harness是一个的终端编程智能体,它要回答的核心问题是:模型何以稳定地交付——目标不漂移、完成有证据、验证有闭环,不说"应该修好了"。为此它在模型与真实世界之间建立一层认知执行环境(CVM),把目标、状态、证据、资源、权限与终止条件从对话历史中外部化,由运行时持续管理。同时对deepseek v4极度优化。
Project memory and workflows for Claude Code and Codex. Keep stories, plans, handovers, and review evidence in your repo. Resume across sessions and follow progress in the Mac app.
Kanwas — Shared context board for teams and agents
CTX Fit finds the cheapest AI coding setup that reliably works on your repository, then applies the winner as a reviewable change.
Composable, view-based memory for DeepSeek Harness. Pluggable sources and strategies, with three-tier memory out of the box.
Save tokens. Maximize context, Safely
Openclaw多智能体协同系统 | Multi-Agent OS for Decision Makers — 基于 OpenClaw (Clawbot) + Slack,让 AI 团队各司其职、自主稳定迭代。
30 sec to give your AI agents persistent memory. Reduce 90% token consumption while also maintaining quality.
Companionship, chat, coding, and work share one memory and context framework — the kind of AI you see in science fiction: it keeps you company, and it gets things done with you.(这是一个基于上下文和注意力机制做的一个多元化的agent项目)
AutoHarness: Automated Harness Engineering for AI Agents
Working memory for Claude Code - persistent context and multi-instance coordination
Context cleaning for Claude Code — prune bloated sessions, protect Agent Teams from context loss, auto-guard with tiered pruning
Open Source Context infrastructure for AI agents. Auto-capture and share your agents' context everywhere.
Local-first AI conversation memory hub to capture, search, summarize, and export chats across major AI platforms. 本地优先的 AI 对话记忆与知识中台。
AI coding agent harness. Five modes: practice to playoffs. Stop your AI from over-engineering. Code like a man. Elbow out bloat. Score clean. ... AI 代理调度框架。五种模式:训练到季后赛。 别让你的 AI 过度设计一切。像个man一样,肘开冗余,干净得分。
Plug-and-play memory for LLMs in 3 lines of code. Add persistent, intelligent, human-like memory and recall to any model in minutes.
AI builder operating system. Turn Claude Code from a per-session tool into a workspace that remembers, maintains itself, and never goes stale. BASE = Builder's Automated State Engine.
The Universal Cross-Model Episodic Memory Standard. Local-first, project-scoped SQLite engine for Google Antigravity, Claude Code, Cursor, Codex CLI, GitHub Copilot CLI, and OpenCode. Zero cloud lock-in.
A complete operating system for Claude that prevents context loss and makes multi-session work reliable
Surgical context engineering with a nice UI, Windows, Mac, Linux, BSD, anywhere where python & Qt run
Tasks, docs, specs and decisions for AI coding agents, as markdown in your repo. Keeps the agent from breaking architectural decisions you already made.
Codex and Claude Code plugin (works in Claude Desktop): choose which past images go to the model with your next message. Unchecked images become placeholders, the conversation goes on, and requests stay small.