MCP server
公式レジストリ、GitHub、npm から 121,886 台のサーバー。カテゴリ、トランスポート、言語で絞り込めます。
公式レジストリ、GitHub、npm から 121,886 台のサーバー。カテゴリ、トランスポート、言語で絞り込めます。
Drive your real, logged-in Chrome from an AI client across multiple profiles with parallel per-tab DevTools control, all locally.
$claude mcp add agent-webbridge -- npx -y agent-webbridgeAllows AI assistants to query web search API benchmarks and perform live search spot checks across providers like You.com, Tavily, Exa, and Parallel.
$claude mcp add websearch-api-eval -- npx -y websearch-api-eval-mcpEnables multi-model AI subagents for Codex and Pi to run parallel tasks with persistent sessions and a local web monitoring panel.
$claude mcp add codex-pi-subagents -- npx -y @caho1/pi-agent-swarmLets an AI assistant read website recordings, replays, heatmaps, funnels, and findings from your Hoot Lens account.
$claude mcp add hootlens-mcp -- npx -y @hootlens/mcpEnables AI agents to programmatically control Burp Suite Professional for automated web security testing, scanning, and exploitation.
Provides self-hosted web search and content extraction for AI agents using parallel multi-engine queries and semantic re-ranking.
$claude mcp add searchpin -- uvx searchpinEnables AI agents to perform multi-source web synthesis with citations using reasoning models and parallel search connectors.
Exposes a GPT-5 inference tool with optional web search and configurable reasoning effort for MCP clients.
Lets an AI agent search the web and fetch page contents so it can find current documentation and verify facts.
$claude mcp add --transport http search-mcp https://search-mcp.parallel.ai/mcpA lightweight Rust-based MCP server that allows multiple AI agents to share a single Chrome instance for parallel web automation and scraping.
Provides a web interface for Claude Code to manage parallel sessions, scheduled jobs, and PR reviews across multiple devices.
Enables AI agents to perform security testing by providing access to over 160 Burp Suite operations and a high-speed Turbo scripting engine.
Enables AI clients to perform parallel browser automation and content extraction through the Zen Browser.
$claude mcp add zenlink-mcp -- uvx zenlink-mcpPerforms orchestrated web research by searching, scraping multiple sources in parallel, synthesizing reports with citations, and fact-checking claims.
Starts deep research jobs and dataset enrichment tasks and retrieves their results as Markdown without leaving the agent.
$claude mcp add --transport http task-mcp https://task-mcp.parallel.ai/mcpEnables AI agents to perform parallel multi-model queries, smart routing, image/video generation, and web search across various providers.
$claude mcp add mcp-multi-model -- npx -y mcp-multi-modelPerforms keyless, multi-engine web research using parallel fanout and browser impersonation to provide token-efficient search results.
Orchestrates multi-agent workflows by delegating heavy coding tasks from manager models to specialized worker agents and rotating Google account quotas.
Runs Playwright MCP browser tools in multiple isolated sessions simultaneously to enable parallel web automation for AI agents.
Performs parallel searches across web engines, Wikipedia, and preprint servers while providing detailed diagnostic receipts for every provider.