MCP1

An MCP server from Zapier that connects authenticated applications so compatible AI tools can access and use them through a single integration layer. It supports clients including Cursor, Claude Code, and Codex.

Awesome MCP Servers is a curated collection of Model Context Protocol (MCP) server implementations, maintained as a GitHub repository with a synchronized web directory. It catalogs production-ready and experimental servers that let AI models interact with local and remote resources through standardized interfaces, including file access, databases, API integrations, browser automation, cloud platforms, research services, and other contextual tools. The repository organizes entries by capability, programming language, deployment scope, operating system, and whether an implementation is official.

Composio MCP Gateway is a governed MCP gateway for connecting AI agents to Composio's managed tools and customer-provided MCP servers. It provides a centralized endpoint through which agents discover and call tools across services such as Slack, Sentry, Datadog, PostHog, and Metabase, instead of connecting separately to each application. The gateway applies organization, team, user, and action-level access policies; keeps credentials out of agent configuration; manages authentication and token refresh; and records tool calls in a shared audit trail. Composio describes it as supporting more than 1,500 managed integrations and production use by enterprise teams.

mcpc is Apify's open-source universal command-line client for the Model Context Protocol (MCP). It maps MCP operations for tools, prompts, resources, asynchronous tasks, skills, notifications, and logging to shell commands, communicating with local stdio servers and remote Streamable HTTP servers. Persistent named sessions keep connections, caches, notifications, and authentication state available across commands. The CLI supports progressive tool discovery through grep and compact tool listings, interactive shells, JSON output for jq, xargs, and shell pipelines, schema validation for scripts and CI, OAuth 2.1 with PKCE and bearer authentication, and a local proxy mode that keeps upstream credentials and OAuth tokens outside AI sandboxes. Its single configuration and credential store can be reused by shell-based AI coding agents, with credentials stored in the operating system keychain. It is distributed as the `@apify/mcpc` npm package.

MCP Inspector is a testing tool for MCP servers. It runs via npx, connects to a local MCP server in a browser, and lets users list and invoke the server's tools.