Actor file specification

{
    // Required field, indicates that this is an Actor definition file and the specific version of the Actor specification.
    "actorSpecification": 1,

    // Required "technical" name of the Actor, must be a DNS hostname-friendly text.
    "name": "google-search-scraper",

    // Human-friendly name and description of the Actor.
    "title": "Google Search Scraper",
    "description": "A 200-char description",

    // Required, indicates the version of the Actor. Since actor.json file is commited to Git, you can have different Actor
    // versions in different branches.
    "version": "0.0",

    // Optional tag that is applied to the builds of this Actor. If omitted, it defaults to "latest".
    "buildTag": "latest",

    // An optional object with environment variables expected by the Actor.
    // Secret values are prefixed by @ and their actual values need to be registered with the CLI, for example:
    // $ apify secrets add mySecretPassword pwd1234
    "environmentVariables": {
        "MYSQL_USER": "my_username",
        "MYSQL_PASSWORD": "@mySecretPassword"
    },

    // Optional field. If true, the Actor indicates it can be run in the Standby mode,
    // to get started and be kept alive by the system to handle incoming HTTP REST requests by the Actor's web server.
    "usesStandbyMode": true,

    // An optional metadata object enabling implementations to pass arbitrary additional properties.
    // The properties and their values must be strings.
    "labels": {
        "something": "bla bla"
    },

    // Optional minimum and maximum memory for running the Actor.
    "minMemoryMbytes": 128,
    "maxMemoryMbytes": 4096,

    // Optional link to the Actor Dockerfile.
    // If omitted, the system looks for "./Dockerfile" or "../Dockerfile"
    "dockerfile": "./Dockerfile",

    // Optional link to the Actor README file in Markdown format.
    // If omitted, the system looks for "./ACTOR.md" and "../README.md"
    "readme": "./README.md",

    // Optional link to the Actor changelog file in Markdown format.
    "changelog": "../../../shared/CHANGELOG.md",

    // Optional link to Actor input or output schema file, or inlined schema object,
    // which is a JSON schema with our extensions. For details see ./INPUT_SCHEMA.md or ./OUTPUT_SCHEMA.md, respectively.
    // BACKWARDS COMPATIBILITY: "inputSchema" used to be called "input", all implementations should support this.
    "inputSchema": "./input_schema.json",
    "outputSchema": "./output_schema.json",

    // Optional path to Dataset or Key-value Store schema file or inlined schema object for the Actor's default dataset or key-value store.
    // For detail, see ./DATASET_SCHEMA.md or ./KEY_VALUE_STORE_SCHEMA.md, respectively.
    // BACKWARDS COMPATIBILITY: "datasetSchema" used to be "storages.keyValueStore" sub-object, all implementations should support this.
    "datasetSchema": "../shared_schemas/generic_dataset_schema.json",
    "keyValueStoreSchema": "./key_value_store_schema.json",

    // Optional path or inlined schema object of the Actor's web server in OpenAPI formation.
    "webServerSchema": "./web_server_openapi.json",

    // Optional URL path and query parameters to the Model Context Protocol (MCP) server exposed by the Actor web server.
    // If present, the system knows the Actor provides an MCP server, which can be used by the platform
    // and integrations to integrate the Actor with various AI/LLM systems.
    "webServerMcpPath": "/mcp?version=2",

    // Scripts can be used by tools like the CLI to do certain actions based on the commands you run.
    // The presence of this object in your Actor config is optional, but we recommend always defining at least the `run` key.
    "scripts": {
        // The `run` script is special - it defines *the* way to run your Actor locally. While tools can decide
        // to implement mechanisms to detect what type of project your Actor is, and how to run it, you can choose to
        // define this as the source of truth.
        //
        // This should be the same command you run as if you were at the root of your Actor when you start it locally.
        // This can be anything from an npm script, as shown below, to a full chain of commands (ex.: `cargo test && cargo run --release`).
        //
        // CLIs may opt to also request this command when initializing a new Actor, or to automatically migrate and add it in the first time
        // you start the Actor locally.
        "run": "npm start"
    }
}