# Apify > Run pre-built browser-automation Actors on managed infrastructure. [Canonical HTML page](https://top-mcps.com/mcp/apify) · [server.json](https://top-mcps.com/mcp/apify.json) · [methodology](https://top-mcps.com/about/methodology) ## Install ### Claude Desktop — `claude_desktop_config.json` Paste under mcpServers. Fully quit and reopen Claude after editing. ```json { "mcpServers": { "apify": { "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### Claude Code — `CLI or .mcp.json` Run from your repo. Commit .mcp.json to share with your team. ```shell # export APIFY_TOKEN=your-apify-api-token claude mcp add apify -- npx -y @apify/actors-mcp-server ``` ### Cursor — `.cursor/mcp.json` Global path: ~/.cursor/mcp.json. Reload window after editing. ```json { "mcpServers": { "apify": { "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### VS Code — `.vscode/mcp.json` VS Code uses the "servers" key (not "mcpServers"). ```jsonc { "servers": { "apify": { "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### Windsurf — `~/.codeium/windsurf/mcp_config.json` Open via Cascade → hammer icon → Configure. ```json { "mcpServers": { "apify": { "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### Cline — `cline_mcp_settings.json` Open via the Cline sidebar → MCP Servers → Edit. ```json { "mcpServers": { "apify": { "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### Continue — `~/.continue/config.json` Continue uses modelContextProtocolServers with a transport block. ```json { "experimental": { "modelContextProtocolServers": [ { "transport": { "type": "stdio", "command": "npx", "args": [ "-y", "@apify/actors-mcp-server" ], "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } ] } } ``` ### Codex CLI — `~/.codex/config.toml` Codex uses TOML. Each server is a [mcp_servers.] subtable. ```shell # ~/.codex/config.toml [mcp_servers.apify] command = "npx" args = [ "-y", "@apify/actors-mcp-server", ] env = { APIFY_TOKEN = "${APIFY_TOKEN}" } ``` ### Zed — `~/.config/zed/settings.json` Zed calls them "context_servers". Settings live-reload on save. ```jsonc { "context_servers": { "apify": { "command": { "path": "npx", "args": [ "-y", "@apify/actors-mcp-server" ] }, "env": { "APIFY_TOKEN": "${APIFY_TOKEN}" } } } } ``` ### ChatGPT — `ChatGPT → Apps directory` Apify doesn't ship a hosted HTTPS endpoint today. ChatGPT supports remote MCP servers only — to use this server in ChatGPT you'll need to deploy it to a public HTTPS URL first (e.g. via Cloudflare Workers or Vercel) or wait for an official remote build. ```none ``` ## At a glance - **Maintainer:** Apify - **Transport:** stdio - **Auth model:** API key - **Required secrets:** APIFY_TOKEN - **Supported clients:** Claude, Cursor, VS Code, Windsurf, Any MCP-compatible client, Apify account - **License:** Apache-2.0 - **Language:** TypeScript - **Latest version:** latest - **Last verified:** 2026-05-27 - **GitHub stars:** 1,275 (fetched 2026-06-02T11:55:52.434Z) - **Score:** 90/100 (rubric 2026-04 — see https://top-mcps.com/about/methodology) - **Source:** https://github.com/apify/actors-mcp-server ## Security & scope - **Access scope:** network - **Sandbox:** Apify API token in env. Actors execute on Apify infrastructure — your code does not run locally. Tokens are scoped per organisation; scope tightly for shared use. - **Gotchas:** - Per-Actor billing means a runaway agent loop can run up bills quickly. Set per-run timeouts. - Public Actors are maintained by third parties — quality varies. Pin to known-good Actors for production. - Scraping is a legal grey zone; some Actors target sites whose terms forbid automation. ## Quick answer **What it does.** Connects to the Apify platform, lists available Actors, runs them with parameters, and returns structured output. Also exposes datasets, key-value stores, and run history. **Best for:** - Web scraping with pre-built Actors - Google Maps / Search / Shopping extraction - Social media scraping (TikTok, Instagram, X) - E-commerce product extraction - Cross-site data extraction at scale **Not for:** - Custom one-off flows with no Actor - Sites with hard anti-bot blocks - Workflows requiring strict cost control (Actors bill per use) ## Description The official Apify MCP gives an AI agent access to the Apify Actor library — thousands of pre-built browser-automation scrapers and workflows running on managed infrastructure. The agent picks an Actor by name (e.g. instagram-scraper, google-maps-extractor), passes parameters, and gets back structured results without spinning up its own browser. ## Why it matters Browser automation has a huge cold-start cost. Apify's Actor library eliminates it — for any common scraping or automation task, an Actor already exists, maintained by someone. ## Key features - Thousands of pre-built Actors - Managed Chromium / proxy infrastructure - Structured dataset output - Run-history + scheduling - Per-Actor pricing transparency ## FAQ ### How does pricing work? Per Actor. Each one publishes its own pricing (per result, per compute unit, or flat). Apify provides $5/month of free platform credit; agent-driven loops can blow through it fast. ### Custom Actors? Yes — write your own Actor in any language and publish it (private or public). The MCP can call private Actors the same as public ones. ### Proxy support? Apify provides residential, datacenter, and Google SERP proxies that Actors use automatically. No separate proxy configuration needed. ### What about TOS / legality of scraping? Your responsibility. Apify operates the infrastructure; you bear the legal exposure for what you scrape. Read site terms and robots.txt before running any agent loop. ## Changelog - **2026-05-27** — Refreshed install snippets and fact sheet; verified for 2026. - **2025-03-25** — Initial directory listing.