Skip to main content
Glama
screaming-frog-mcp

screaming-frog-mcp

screaming-frog-mcp

Servidor MCP que permite a Claude ejecutar rastreos headless de Screaming Frog SEO Spider, exportar datos y gestionar el almacenamiento de rastreos, sin necesidad de abrir la interfaz gráfica.

Escribe una URL en Claude. Screaming Frog se ejecuta en segundo plano. Obtienes los datos de vuelta. Eso es todo.

Bifurcado de bzsasson/screaming-frog-mcp v0.1.0 con correcciones de errores. El original tenía problemas que lo hacían inutilizable en la práctica: bloqueos de tuberías (pipe deadlocks) que colgaban los rastreos, detección falsa de GUI que bloqueaba todo después de la primera ejecución, un comando de eliminación que podía borrar toda tu base de datos de rastreos. Todo solucionado.


Qué se ha corregido

Error

Solución

Bloqueo de tubería

stdout/stderr redirigidos a archivos de registro en lugar de PIPE. Los rastreos ya no se cuelgan cuando SF produce una salida grande.

Detección de GUI

Utiliza psutil en lugar de ps aux. Funciona en Mac y Windows. Los procesos CLI headless ya no se confunden con la GUI.

Limpieza de rastreos obsoletos

SF deja un archivo temporal crawl.seospider dentro de su propio paquete de aplicación cuando se interrumpe un rastreo. Cada rastreo posterior falla. Ahora se limpia automáticamente antes de cada ejecución.

Seguridad de eliminación

delete_crawl(".") solía resolverse en el directorio de datos raíz y borrar todo. Solucionado.

Fuga de directorio de exportación

Las exportaciones fallidas dejaban directorios temporales en el disco. Ahora se limpian.

Validación de entrada

Listas de permitidos de caracteres más estrictas para argumentos CLI y db_id.


Related MCP server: Screaming Frog SEO Spider MCP Server

Requisitos


Instalación

Mac

uvx --from git+https://github.com/marykovziridze/screaming-frog-mcp screaming-frog-mcp

Añadir a ~/Library/Application Support/Claude/claude_desktop_config.json:

{
  "mcpServers": {
    "screaming-frog": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/marykovziridze/screaming-frog-mcp", "screaming-frog-mcp"]
    }
  }
}

Windows

Instalar uv primero:

powershell -ExecutionPolicy ByPass -c "irm https://astral.sh/uv/install.ps1 | iex"

Añadir a C:\Users\[nombre]\AppData\Roaming\Claude\claude_desktop_config.json:

{
  "mcpServers": {
    "screaming-frog": {
      "command": "uvx",
      "args": ["--from", "git+https://github.com/marykovziridze/screaming-frog-mcp", "screaming-frog-mcp"],
      "env": {
        "SF_CLI_PATH": "C:\\Program Files (x86)\\Screaming Frog SEO Spider\\ScreamingFrogSEOSpiderCli.exe"
      }
    }
  }
}

Reinicia Claude Desktop después de editar la configuración.


Herramientas

Herramienta

Qué hace

sf_check

Verifica que SF esté instalado y licenciado

crawl_site

Inicia un rastreo headless

crawl_status

Comprueba el progreso del rastreo

list_crawls

Lista los rastreos guardados en la base de datos de SF

export_crawl

Exporta los datos del rastreo como CSV

read_crawl_data

Lee y filtra los datos CSV exportados

delete_crawl

Elimina un rastreo guardado

storage_summary

Muestra el uso de disco del almacenamiento de rastreos


Configuración

Variable

Predeterminado

Notas

SF_CLI_PATH

Mac: autodetectado

Configurar manualmente en Windows o instalaciones personalizadas


Limitaciones conocidas

  • Ruta de rastreo obsoleto en Windows — la limpieza automática funciona en Mac. En Windows, si los rastreos fallan después de una interrupción, busca un archivo crawl.seospider en tu directorio de instalación de SF y bórralo manualmente.

  • Sin porcentaje de progreso de rastreo — la CLI headless de SF no informa del progreso a mitad del rastreo. Sabes cuándo empieza y cuándo termina.

  • Sitios grandes — probado en sitios de hasta ~160 páginas. No probado bajo estrés en sitios de más de 10.000 páginas.


Licencia

MIT — ver LICENSE


Créditos

Servidor MCP original de Boaz Sasson.

Available Tools

8 tools
crawl_siteA

Start a background Screaming Frog crawl that saves to SF's internal database.

Args: url: The URL to crawl (e.g. https://example.com) config_file: Optional path to a .seospiderconfig file for crawl settings (including crawl limits) label: Optional label for identifying this crawl (e.g. 'freshgovjobs')

Returns: A crawl_id to use with crawl_status to check progress. The crawl runs in the background - use crawl_status to poll.

Note: To limit the number of URLs crawled, export a .seospiderconfig from the SF GUI with the desired crawl limit, then pass it via config_file.

Inputs: url (required), config_file, label. Returns: result

crawl_statusC

Check the status of a running or completed crawl.

Args: crawl_id: The crawl_id returned by crawl_site

Inputs: crawl_id (required). Returns: result

delete_crawlA

Delete a crawl from Screaming Frog's internal database to free disk space.

Args: db_id: The Database ID from list_crawls

WARNING: This permanently deletes the crawl data. It cannot be undone.

Inputs: db_id (required). Returns: result

export_crawlA

Load a saved crawl from SF's database and export data as CSV files.

Args: db_id: The Database ID from list_crawls (e.g. '1234' or a crawl identifier) export_tabs: Comma-separated export tabs (default: Internal:All,Response Codes:All,Page Titles:All,Meta Description:All,H1:All,H2:All,Images:All,Canonicals:All,Directives:All). See the export-reference resource for all options. bulk_export: Optional bulk export types (e.g. 'All Inlinks,All Outlinks') save_report: Optional reports to save (e.g. 'Crawl Overview')

Returns: An export_id and list of generated CSV files. Use read_crawl_data to read them.

Inputs: db_id (required), export_tabs, bulk_export, save_report. Returns: result

list_crawlsA

List all crawls saved in Screaming Frog's internal database. Returns crawl names, Database IDs, and sizes. Use the Database ID with export_crawl or delete_crawl.

Inputs: no named parameters. Returns: result

read_crawl_dataA

Read CSV data from an export. Use after export_crawl.

Args: export_id: The export_id from export_crawl file: CSV filename to read (from the file list in export_crawl output) limit: Max rows to return (default 100) offset: Number of rows to skip (for pagination) filter_column: Optional column name to filter by filter_value: Optional value to match in the filter column (case-insensitive substring)

Returns: CSV data as formatted text with column headers.

Inputs: export_id (required), file (required), limit, offset, +2 more. Returns: result

sf_checkA

Verify that Screaming Frog SEO Spider is installed and the CLI is accessible. Returns version info and license status.

Inputs: no named parameters. Returns: result

storage_summaryA

Show disk usage of Screaming Frog's internal crawl storage. Returns total size and per-crawl breakdown of ProjectInstanceData.

Inputs: no named parameters. Returns: result

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

Loading tool schema changelog…

    TDQS

    A4.2/5.0

    Scored across 8 tools

    Disambiguation5/5

    Each tool has a distinct purpose with no overlap: crawl_site initiates crawls, crawl_status checks progress, delete_crawl removes data, export_crawl exports data, list_crawls lists saved crawls, read_crawl_data reads exported data, sf_check verifies installation, and storage_summary shows disk usage. The descriptions clearly differentiate their functions, preventing agent misselection.

    Naming Consistency5/5

    All tools follow a consistent snake_case naming pattern with clear verb_noun structures (e.g., crawl_site, list_crawls, read_crawl_data). The naming is predictable and readable throughout the set, with no deviations in style or convention.

    Tool Count5/5

    With 8 tools, this server is well-scoped for managing Screaming Frog crawls. Each tool earns its place by covering essential operations like starting, monitoring, exporting, and managing crawls, without being overly sparse or bloated. The count aligns perfectly with the domain's workflow needs.

    Completeness5/5

    The tool set provides complete CRUD/lifecycle coverage for Screaming Frog crawls: create (crawl_site), read (export_crawl, read_crawl_data), update (implied via crawl_status monitoring), delete (delete_crawl), and management (list_crawls, storage_summary, sf_check). There are no obvious gaps, and agents can handle all core tasks without dead ends.

    Related MCP Connectors

    Related MCP Servers