Hierarchical RAG architecture scaling to 693K chunks on consumer hardware (4GB VRAM). Features 3-address routing, hybrid vector+graph fusion, and SetFit classification.

🧠 Persistent memory MCP for AI agents — Knowledge graph + Hebbian learning + Anti-hallucination. 12 tools, 1 dependency, zero manual setup.

Reliable research infrastructure for AI agents. Evidence-backed web search with citations, confidence scores, and Clarity anti-hallucination. MCP server, REST API, Python SDK.

EN: An overfitted SD prompt engine with severe "aesthetic snobbery," forcibly transforming mundane ideas into professional-grade physical rendering instructions. CN: 一个具备“审美洁癖”的过拟合提示词引擎，强行将平庸构思纠偏为具备极致物理质感的工业级渲染指令。

Free web search for OpenClaw: self-hosted SearXNG + Scrapling anti-bot + multi-source cross-validation. No API key. No cost. Tells you how much to trust the answer.

A strict, deterministic LLM protocol for loading, reading and activating the DCQN.MATRIX Axiomatics from the OSF DOI (10.17605/OSF.IO/QWA6S), including anti-simulation safeguards and full formal reconstruction into DCQN_LOGIK_SESSION_V1.

65 plugins that turn Claude Code into an autonomous development team. 24 agents, 34 skills, 5 hooks. Includes 12-plugin anti-hallucination suite. One-line install.

The Anti-Hallucination data layer for B2B Sourcing. Deep-verified global supply chain entities designed for RAG and LLM instruction tuning.

🌀 Verifiable multi-agent AI with shadow auditing that transforms uncertain scenarios into transparent, confidence-scored strategic decisions.

I built a production-style RAG system focused on grounded generation, not open-ended LLM output. Design priorities: retrieval quality, validation, and measurable confidence not just document chat.

RAG-powered LLM assistant for HR policy Q&A with ChromaDB, guardrails, citation tracking, and evaluation framework. FastAPI + Streamlit.

Evaluate anything. Predict everything. Hallucinate nothing. — SNN-verified document evaluation & prediction credibility MCP server.

AI-driven software development methodology, anti-hallucination pipeline with specialized agents, strict gates, and adversarial review.

openclaw-Grounding Guard：有效抑制 OpenClaw 幻觉率（自动注入可追溯上下文 + 出站标注审计告警）

Cryptographic protocol for publishing machine-verifiable facts about organizations. Ed25519 signed claims for AI systems.

MCP server that gives AI coding assistants persistent memory, context control, and anti-hallucination tools

A local-first agentic RAG system demonstrating temporal reasoning. Uses Graphiti that grounds its responses in an evolving Neo4j knowledge graph and Pydantic AI with local Ollama models for strict, fact-based retrieval to eliminate AI hallucinations.

ATUM SYSTEM — Claude Code plugin: 70 agents, 167 skills, 81 commands, 37 rules, 36 hooks, 34 MCP servers. Install: claude plugins install arnwaldn/atum-system

The Open Protocol for AI-Agent Sourcing. Stop LLM hallucinations in B2B. | AI 时代企业可信数字主权生成引擎。

A safe, transparent, and human-controlled self-evolution framework for AI agents. Three-zone safety architecture ensures AI learns autonomously while humans stay in control.