Search

Terminal AI Agent Library — TAP IR + Model-Specific Compilation

AI-agent-friendly tmux wrapper: unified local + remote sessions, single-pane sessions, safe primitives.

Corpus-coupled, drift-free routing for multi-collection RAG.

Multi-agent framework with 684 skills, constitutional governance, 19-layer runtime (pipeline, privacy, AI, skills, cognitive, scalability, observability), and Anthropic SDK integration.

Universal AI Agent node for n8n — supports ALL chat models via OpenAI-compatible fallback. Drop-in replacement for Better AI Agent.

A Unified Virtual Filesystem For AI Agents

Brain-inspired cognitive memory database for LLMs

LangChain connector for Noxy human-in-the-loop — routes to all registered devices via agent middleware and webhooks.

AI-driven functional testing agent. Point it at a URL; it explores the app, generates a test plan, runs it, and reports findings. Web (Playwright) + visual regression + accessibility (axe-core) + REST API contract tests, all driven by your choice of LLM.

AgentGuard Spend plugin for Hermes Agent (Nous Research). One-step install of local spend caps + signed audit logs for every Anthropic/OpenAI call your Hermes agent makes. No proxy, prompts never leave your machine.

Docker Compose for AI workflows. Define AI agents, RAG pipelines, MCP servers, and multi-LLM apps in a single YAML file — a lightweight alternative to LangChain.

n8n community node: Anthropic Claude chat model via Google Cloud Vertex AI (Model Garden). Wraps @langchain/anthropic with @anthropic-ai/vertex-sdk so Claude models hosted on Vertex AI (e.g. claude-haiku-4-5@20251001 in europe-west1) can be used as a Chat

Multi-provider AI agents SDK for SvelteKit + Node. Thin agent/tool/run/runStream surface over the Vercel AI SDK — OpenAI, Anthropic, Google Gemini, Mistral, Cohere. Zod-typed tools, full streaming event surface, optional HTML sanitizer.

Add-First Agent skeleton — bring your own LLM, bring your own tools

WuYa (无涯) — A theory-driven multi-agent system for academic paper evaluation and journal recommendation

Hard budget limits and guardrails for LLM APIs (OpenAI, Anthropic, Google, Mistral).

Tuple-Space + Yool safe-speed runtime kernel: lazy 1,000,000+ subagent batch_spawn, adaptive lane concurrency, and a dependency-free OpenAI-compatible provider client for real subagents on DeepSeek / MiMo / OpenRouter / local LLMs.

Claude Patcher -- MCP tools that extend Claude Code beyond built-in capabilities. Fix encoding, convert formats, detect duplicates, batch rename, archive, and more.

**Secure AI execution framework for Web3 agents.**

n8n community nodes for TestMu AI Browser Cloud (formerly LambdaTest). Run browser automation scripts on a real cloud browser, or give an AI Agent native browser tools — navigate, click, type, extract — over Playwright on TestMu's cloud infrastructure.