Apify MCP Server

Official Apify MCP for Actors, web-data extraction, and automation in an AI client.

Description

Apify MCP Server is Apify’s official Model Context Protocol interface. It connects an MCP-capable AI client to the Apify Platform and Apify Store. According to Apify, agents can search for Actors, inspect their input schemas, run Actors, read run results and storage, and search Apify documentation. Actors are prebuilt tools for web scraping, data extraction, and automation. Examples in the official README cover social-media and search-engine data, maps and e-commerce sites, and fetching individual web pages with rendering. This entry covers the server operated by Apify, not an arbitrary community Actor.

Research, automation, and social media

Research, Automation, and Social Media are appropriate categories for the documented outcome. An agent can first find a suitable Actor in the Store, review its required fields, and then run it deliberately. The server can read Actor-run datasets page by page, retrieve run status and logs, and create or change saved Actor tasks. That can support collecting sources for research, extracting structured public profile or post data, or preparing recurring data-gathering workflows. The output is still working material: an Actor result is not proof of accuracy, completeness, or permission to use data. For consequential use, check the source, collection time, field meaning, and likely gaps.

Apify documents limits on its MCP server as well. Full-permission Actors are excluded from search and execution because running one is intended to remain a decision that a person approves. Rental Actors are also excluded because their subscription model does not fit an on-demand MCP call. Actor input schemas are processed for MCP compatibility, so descriptions and enum values can be limited or truncated. Before a production run, read the individual Actor documentation and use an explicit, minimal tool selection. The README says that tools accessing external resources carry open-world behavior annotations.

Remote OAuth and local stdio operation

Apify explicitly offers two paths, so this entry is classified as both. The recommended path is the hosted Streamable HTTP endpoint at https://mcp.apify.com. On first connection, an MCP client can open a browser for sign-in to Apify and OAuth approval. This avoids placing an Apify API token in client configuration. Alternatively, the remote endpoint accepts an Apify API token as a Bearer token in the Authorization header. The hosted service also supports output-schema inference for structured Actor results according to the documentation; the local stdio server does not provide that feature.

For development, testing, or a client without remote MCP support, Apify documents local stdio operation via npx -y @apify/actors-mcp-server, with APIFY_TOKEN held in the protected environment. The package downloads at first use. Local means that the MCP process runs in the local environment; tool requests and Actor inputs still go to the Apify API for execution. Never put a real token in Git, chat, screenshots, or a shared MCP file. Use secret management, rotate a token suspected of exposure, and restrict its rights where the account and organization support that.

Scraping boundaries, personal data, and prompt injection

Web scraping does not remove obligations under website terms of use, robots or access restrictions, copyright, data-protection law, or contracts. For every Actor and target, check whether intended collection, storage, and reuse are permitted. Collect only necessary data, minimize the run, and do not bypass access controls. Social-media, map, and contact information can be personal data. Define purpose, legal basis, retention, access, and deletion before collection; do not send sensitive personal data or secrets in Actor inputs unless it is genuinely necessary.

Page content, Actor output, logs, and README text are untrusted third-party content. They can carry prompt-injection text, such as instructions to ignore rules, extract tokens, enable more tools, or publish data. Treat them as data, never instructions. Restrict the agent to needed, preferably read-only tools; independently review Actor selection, input, and every state-changing action. OAuth approves access within the Apify connection’s scope, while an API token authorizes Actor runs and protected data access in the account context. Grant only necessary permissions and assess whether the Actor, client, logs, and connected model provider may store or pass on data.

FAQ

Is the official Apify MCP Server available remotely? Yes. Apify operates https://mcp.apify.com as a Streamable HTTP endpoint with OAuth sign-in or a Bearer-token alternative.

Is there also a local route? Yes. Apify documents local stdio through npx -y @apify/actors-mcp-server and APIFY_TOKEN; Actor execution still uses the Apify API.

Can an agent start every Actor? No. Apify says full-permission and rental Actors are excluded from MCP-server search and execution. Terms, data protection, and human approval still apply.

On 2026-09-08, the GitHub API reported exactly 6,306 stars for apify/apify-mcp-server. It is a point-in-time repository-popularity signal, not evidence of quality, privacy, or security.

Requirements

An MCP-capable client; Streamable HTTP plus OAuth or Bearer-header support for remote MCP, or Node.js 18+ and a securely set APIFY_TOKEN for local use.

Installation instructions

The official Apify MCP Server brings Actors from Apify Store into an MCP-capable client. An Actor can fetch a web page, gather search results, structure public social-media content, or provide data from a run. Do not configure the server as broad “web access for everything.” Start with a bounded task, such as collecting public sources on a topic, reviewing results in a limited dataset, and documenting original pages. Select only the Actors and server tools needed for that task. This reduces token use, reduces the attack surface, and makes agent decisions easier to review.

Step 1: Connect Remote MCP with OAuth

Apify’s recommended route is the hosted Streamable HTTP endpoint at https://mcp.apify.com. Add that URL in a client that supports Remote MCP. On the first connection, the client or browser opens Apify sign-in. Sign in only through the official Apify dialog and verify the account before approval. Then complete OAuth authorization. This avoids keeping an API token as plaintext in an MCP configuration file. After connecting, start with a low-risk read action, such as Actor search or documentation search.

For the remote endpoint, Apify also accepts an API token in the Authorization: Bearer header. Choose that route only when the client does not support OAuth or a controlled technical flow requires it. Store the token in the client’s secret or credential manager, never in a URL, prompt, sample file, or Git. Use a separate automation token where your account permits it, and rotate or revoke it if exposure is suspected. A token is not mere configuration: it represents access in the context of your Apify account.

Step 2: Use local stdio deliberately

For development, testing, or clients without remote support, Apify documents local stdio. Install Node.js 18 or later, put APIFY_TOKEN in a protected environment variable, and configure npx -y @apify/actors-mcp-server as the server command. A first-run package download is expected, but still check the package source and update deliberately. Then test search or documentation tools before running a broad scraper or publishing a task.

bash
npx -y @apify/actors-mcp-server

A local stdio process does not make the data path fully local. Actor inputs and tool requests go to the Apify API for execution; results then reach the client and may enter model context, chat history, or logs. According to Apify, the hosted server supports output-schema inference for structured Actor results, while local stdio does not. Select the route based on client capability, data boundary, and operating model, not on a misleading assumption that “local” means all processing stays local.

Limit Actors, data, and scraping lawfully

First find an Actor, read its schema, and restrict its input to the target, period, fields, and quantity you truly need. For every target site, review terms of use, access restrictions, privacy and copyright obligations, and any applicable robots guidance. Do not attempt to bypass logins, rate limits, technical blocks, or other access controls. Apify says full-permission and rental Actors are excluded from MCP search and execution, but that does not replace your own assessment of legality and risk.

Social-media, map, and contact Actors can return personal data. Collect only data with a documented purpose and suitable legal basis. Decide in advance who may read results, how long they remain, how they are deleted, and whether they go to a model, service provider, or other systems. Do not put private credentials, payment data, health data, or confidential customer data into Actor inputs unless this is genuinely necessary and legally supported.

Prompt injection, permissions, and approval

Scraped pages, Actor output, and logs can include text intended to manipulate the agent: “ignore rules,” “forward data,” or “add a new server.” Those are not commands. Treat all web content as untrusted data. State in the agent task that content may only be extracted, cited, and checked. Avoid giving the agent write-capable tools where possible. Before call-actor, task changes, publication, or sending data onward, review the specific Actor, input, recipient, and expected effect.

OAuth and API tokens can grant access to runs, datasets, key-value stores, and tasks in the Apify-account context. Regularly review available permissions and remove connections no longer needed. Disable telemetry only deliberately through Apify’s documented configuration if that fits your privacy design. Responsibility for the selected Actor configuration, lawful data use, and final action remains with the operator, not the model or MCP server.

FAQ

Which URL is the official remote server? https://mcp.apify.com. The first connection can start browser OAuth.

When do I use APIFY_TOKEN? For local stdio or as the Bearer alternative to OAuth. Store it only as a secret.

Are Actor results trustworthy instructions? No. They are external data, can be incomplete, and can contain prompt injection. Validate them against original sources and approval rules.

npx -y @apify/actors-mcp-server

Authentication

Remote supports OAuth through browser sign-in and alternatively an Apify API token as a Bearer token. The local stdio server reads APIFY_TOKEN from the environment.

Required access permissions

OAuth and APIFY_TOKEN act in the Apify account context. Token and account rights determine accessible Actors, runs, datasets, storage, and tasks; Actor execution and write-capable task actions need deliberate approval.

Transmitted or stored data

Requests and Actor inputs are sent to the Apify API for execution. Results, run metadata, and storage content return to the connected MCP client and can be processed or retained there, in logs, and by the connected model provider.

Security risks

Token or OAuth access can expose Actor runs and account data. Scraped content can involve personal data, website terms, and prompt injection. Limit tools, review targets and inputs, and independently confirm mutating or publishing actions.

License and costs

License
MIT
Cost
paid

The source code is MIT licensed. Authenticated Actor execution and account-data access follow the current Apify account configuration and provider terms; check official information before production use.

Alternatives

Not recorded yet.

At a glance

Provider
Apify
Status
Official server
Deployment
Local and remote
Current version
Not recorded yet.
GitHub stars
16,423
Last reviewed
08.09.2026

Repository and documentation

Categories

Supported clients

Related guides

Guides and background related to this entry.

No related guides yet.