Search

EdgeCrab 🦀 A Super Powerful Personal Assistant inspired by NousHermes and OpenClaw — Rust-native, blazing-fast terminal UI, ReAct tool loop, multi-provider LLM support, ACP protocol, gateway adapters, and built-in security hardening.

Always-on JARVIS-style personal assistant: a Claude tool-use agent (Python daemon) that controls Home Assistant, Spotify, Android TV, and your PC by voice or text — with tiered LLM routing, a hard monthly budget cap, and a security guard.

Ardur — open-source personal AI agent. Lives in your terminal and across your messaging channels. Multi-provider LLM access, tool use, memory, plugins, with cap-token security, receipt-chain transparency, and cost-tuple awareness built in.

AgentBridge allows you to instantly connect any website to AI agents like Claude, Cursor, and Windsurf. It automatically crawls the web and generates a secure MCP server, giving your LLMs native control to read pages and perform authenticated actions via your browser.

AI-CLI: an AI-powered automated security orchestration engine and CLI tool built in Go for Termux & Kali Linux. Driven by Gemini API with a strict 5-stage tactical framework, live dashboard tracking, and an automated self-updater

Polyglot AI agent orchestration core (Rust reference + TS/Go/.NET/Python bindings) — agents, workflows, tools, checkpointing, memory, HITL, cost. A security-first LangGraph; the engine behind smooth-operator and lom.smoo.ai.

Built on a blazing-fast Rust core, seamlessly unifying relational databases, NoSQL, vector stores, analytics engines, Serverless, and local files. With native AI Agent, secure terminal, and 20+ productivity tools — ending the era of juggling multiple clients.

Distributed compute intelligence network with WASM cryptographic security - contribute browser compute, spawn distributed AI agents, earn credits. Features Ed25519 signing, PiKey identity, Time Crystal coordination, Neural DAG attention, P2P swarm intelli

The Straw Hat crew of AI agents and skills: brainstorm, plan, execute, checkpoint, quality, gates, review, security, self-healing. Installs into Claude Code, opencode, Copilot, Gemini, Codex, Cursor, Kimi, pi, Windsurf, Cline, Kilo, Antigravity.

An AI-powered coding assistant built with Python, OpenRouter, and the OpenAI SDK. Demonstrates secure tool calling, file system operations, Python execution, and iterative agent workflows inspired by modern AI coding tools like Claude Code and Cursor.

OpenBee — Busy as a Bee, Smart as AI, Automated for You. Building a user-friendly, strong, and secure AI agent/skills CLI tool. A hive of specialized AI bees working in harmony, where every role contributes unique strengths and distinct capacities to make

Autonomous multi-agent lead enrichment and scoring system built with Google ADK 2.0, utilizing a local MCP server for data enrichment, a security checkpoint node for prompt injection and PII protection, and human-in-the-loop (HITL) manual reviews.

React Doctor is an MIT-licensed React code quality scanner for teams using coding agents and modern frontend stacks. It runs from the CLI or CI, returns a health score, and reports diagnostics across React patterns, performance, architecture, security, and accessibility.

Turn your AI coding assistant into a DevOps-specialized agent using the CloudBees Unify MCP server. 7 ready-to-use skills for pipeline visibility, build triage, security scanning, release readiness, feature flags, CI health, and Jira ticketing.

Venus CLI is a modular, extensible, and secure AI Agent command-line interface. It is built from scratch using Node.js and TypeScript, designed not just as a chatbot, but as a fully capable agent that can read/write files, analyze projects, execute comman

On-device MCP server for Android. Connect ChatGPT through OpenAI's official secure tunnel, let AI agents operate your phone and run scheduled automations on it — App edition (Accessibility/Shizuku) or root edition (Magisk/KernelSU/APatch). No PC, no ADB. 卓爱桥

Autonomous multi-agent lead enrichment and scoring system built with Google ADK 2.0, utilizing a local MCP server for data enrichment, a security checkpoint node for prompt injection and PII protection, and human-in-the-loop (HITL) manual reviews.

FastMCP-based SSH bridge enabling AI assistants (Claude, ChatGPT, Copilot) to securely execute commands on remote servers. Supports STDIO & HTTP/SSE transports, OAuth 2.0/OIDC authentication, multi-host orchestration, and persistent SSH sessions.

Platform for storing, versioning, and managing the operational knowledge of AI systems—including agent plans, guides, styles, and more—centralized on a scalable backend powered by R2 or S3. Through an MCP server, agents can securely create, update, and retrieve their own execution context in a structured way.

Cognitive OS v3.0 — Universal AI agent operating system specification. 16-section cognitive architecture for Claude Code, Cursor, Codex, Gemini, and Factory Droid. Multi-harness prompt engineering with verification chain, swarm coordination, and adversarial security.