Agent Base is a source-level research project on coding agents. It compares Codex CLI, OpenCode, Gemini CLI, Kimi CLI, and SWE-agent across agent loops, tools, MCP integration, context/memory handling, UI flows, web architecture, and safety controls.

A practical, community-driven checklist for pentesting MCP servers. Covers traffic analysis, tool-call behavior, namespace abuse, auth flows, and remote server risks. Maintained by Appsecco and licensed for remixing.

Official repository for the paper, "MCP-Flow: Facilitating LLM Agents to Master Real-World, Diverse and Scaling MCP Tools"

The intelligent MCP server for AI-assisted Godot 4 development. 35 tools for spatial intelligence, code understanding, flow tracing, and visual debugging.

This project demonstrates how to build a multi-agent AI automation framework using the Model Context Protocol (MCP). The setup enables LLMs (Claude AI in this case) to autonomously execute UI flows, API validations, file operations, and cross-system authentication workflows through standardized tool interfaces

An editable MCP tool exposing 'YOU' a human as a tool the AI can call for discussion mid flow.

Node.js/TypeScript MCP server for AWS Single Sign-On (SSO). Enables AI systems (LLMs) with tools to initiate SSO login (device auth flow), list accounts/roles, and securely execute AWS CLI commands using temporary credentials. Streamlines AI interaction with AWS resources.

Secure-Hulk is a security scanner for Model Context Protocol (MCP) servers and tools. It helps identify potential security vulnerabilities in MCP configurations, such as prompt injection, tool poisoning, cross-origin escalation, data exfiltration, and toxic agent flows.

Flow blockchain tools for Model Context Protocol (MCP)

Deterministic MCP tool flows for AI agents. Remove unnecessary LLM calls between predictable tool steps.

Taias helps you turn your MCP server's pile of tools into opinionated flows

MCP Demos - 8 Node.js/TypeScript demos showcasing Model Context Protocol capabilities, from text-only tools to interactive UIs with bidirectional communication. Covers elicitation, OAuth consent flows, async progress tracking, cancel/retry, MCP App resources (ui://), and report export. No external dependencies required. Includes 5-minute and 15-min

A lightweight, high-performance MCP server connecting AI, tools, and automation through Stdio and HTTP streamable transports — enabling real-time context flow, smart orchestration, and seamless collaboration.

End-to-end demo of the Model Context Protocol (MCP) with FastAPI, showing how browser input flows through MCP Client & Server tools to structured health-related results.

MCP server for Shopify with full OAuth flow, 30+ tools for products, collections, inventory, draft orders, metafields, and bulk operations.

This repository contain sample code of ADK agent app that can handle flow of multimodal request using custom tools and mcp server

Norce Commerce Agent SDK: open-source backend + embeddable UI widget for building conversational shopping assistants on top of Norce Commerce via the Norce MCP Server. Includes an Agent BFF, tool integrations, and reference implementations to help partners add agentic product discovery and cart flows to webshops.

MCP server for managing webMethods Integration Server via pure HTTP API - 39 tools for flow services, adapters, ports, and more

MCP server for Kentik network observability — 8 tools covering flow queries, DDoS alerting, BGP routing, synthetic monitoring, and AI-powered analysis

🔍 Inside Claude Code — Architecture Deep Dive | Flow Diagrams, UML, Sequence Diagrams | How Memory Shrinking, Tool Execution, Permissions & MCP Work Under the Hood

Salesforce Meta Tool (MCP Server) with end-to-end identity propagation via On-Behalf-Of (JWT Bearer) flow through Azure AI Foundry and APIM

Implements Anthropic's "Code Execution with MCP" pattern for building more efficient AI agents. Enables executing TypeScript code that calls MCP tools, achieving 97%+ token savings via progressive tool disclosure, context-efficient results, powerful control flow with loops and conditionals, state persistence, and a reusable skills system.

Stateful MCP server that surfaces tools by task context, shrinking choices to ≤5 per workflow. Demonstrates notes and MLOps flows with progress streaming, Pydantic metrics, and clean session resets.

A Python-based server implementation of the Model Context Protocol (MCP) designed for Retrieval-Augmented Generation (RAG) workflows. It manages the flow of context, tool invocation and model output in one unified service. Supports modular “Models” and “Modules” directories for extensible integrations, environment configuration, and easy deployment

Complete MCP Tool Flow Architecture - OAuth Gateway, Semantic Tool Filter, Name Bridge, Context Forge aggregation. 95% token reduction for Claude.ai integration.

An educational tool that visualizes Model Context Protocol (MCP) communication between an LLM-powered chat interface and MCP servers, designed to teach users how MCP orchestrates tool calling through transparent, real-time message flow visualization.

MCP server for Node-RED — safe deploy with optimistic locking, smart search, 13 tools for AI-driven flow management

56-tool MCP server for Aster DEX perpetual futures — VPVR analysis, Binance flow signals, regime detection, and automated strategies

demonstrating Model Context Protocol (MCP) authorization with a minimal, working server implementation. Shows how tools, authorization flows, and server-client interactions work inside a VS Code MCP support.

The most complete MCP server for API testing. 27 tools: requests, assertions, flows, OpenAPI, mock data, load testing, collections, environments, cURL export, response diffing. Zero config, zero dependencies.