Glama MCP Registry Scraper

🔎 Glama MCP registry scraper for server data

Glama MCP Registry Scraper collects public server listings from the Glama MCP registry. Use this glama mcp registry scraper to export Model Context Protocol server metadata from the catalog, a search query, or exact Glama server IDs without opening each listing by hand.

Each saved row is one Glama MCP server. Rows include the server ID, name, namespace, slug, description, Glama attributes, source repository URL, license, environment schema, required and optional environment variables, hosting readiness, repository language, star count, homepage, source update timestamp, and source-backed tool records when Glama exposes them.

You do not need a Glama account, API key, cookies, or private workspace access. For a quick first run, keep the prefilled github search query and Maximum servers set to 144. The run saves clean dataset rows you can export from Apify, query through the API, schedule, or send to another tool through an integration.

✅ What this Actor does

• Scrapes the Glama MCP catalog and saves one row per MCP server.
• Searches server names, descriptions, namespaces, and attributes with one or more search queries.
• Looks up exact Glama server IDs for detail runs.
• Filters catalog or search output by Glama attributes such as hosting:remote-capable or author:official.
• Filters by repository language when the public source repository exposes it.
• Extracts environment schema facts into requiredVariables, optionalVariables, and the original source schema.
• Adds mechanical readiness fields from source-backed data, such as hosting type, environment requirements, and repository metadata.
• Charges only successfully saved server rows.

📦 Data you can export

The dataset is built for developers, AI tooling teams, registry builders, and product researchers who need MCP server data in a repeatable format.

Typical fields include:

• serverId, name, namespace, and slug
• description and source-native attributes
• readiness.hosting, readiness.score, and readiness.reasons
• environment.requiredVariables, environment.optionalVariables, and environment.schema
• sourceRepositoryUrl, homepageUrl, license, language, starCount, and sourceUpdatedAt
• tools, resources, and prompts arrays when source-backed records are exposed

Glama server page URLs are not duplicated in the dataset because they can be rebuilt from serverId as https://glama.ai/mcp/servers/{serverId}.

🚀 How to run it

Fill one target section per run:

1. Use Search queries for the easiest first run. Example: github, slack, or database.
2. Use Collect catalog to export the current Glama MCP catalog.
3. Use Server IDs when you already know exact Glama IDs, such as hl7cazcjt9.

Then set Maximum servers. Attribute filters and language filters apply to catalog and search runs. If you fill more than one target section, the Actor stops safely and asks you to clear the extra target sections.

🧾 Input reference

• collectCatalog: Collect matching servers from the Glama catalog.
• searchQueries: Search Glama MCP servers by names, descriptions, namespaces, and attributes.
• serverIds: Look up exact Glama server IDs.
• attributeFilters: Keep only rows with matching source attributes.
• languageFilters: Keep only rows with matching repository language.
• maxItems: Limit how many matching servers are saved.

🧪 Output example

{
  "serverId": "hl7cazcjt9",
  "name": "proofdrop-mcp",
  "namespace": "mezotv",
  "slug": "proofdrop",
  "description": "Enables AI agents to upload temporary public review artifacts...",
  "attributes": ["hosting:remote-capable"],
  "readiness": {
    "hosting": "remote-capable",
    "score": 90,
    "reasons": [
      "Remote-Capable hosting",
      "Environment schema exposes required variables",
      "Repository language is available",
      "Repository update timestamp is available"
    ]
  },
  "tools": [],
  "resources": [],
  "prompts": [],
  "installCommand": null,
  "environment": {
    "requiredVariables": ["AWS_REGION", "AWS_ACCESS_KEY_ID"],
    "optionalVariables": ["MCP_PORT"],
    "schema": {
      "type": "object",
      "required": ["AWS_REGION", "AWS_ACCESS_KEY_ID"]
    }
  },
  "sourceRepositoryUrl": "https://github.com/mezotv/proofdrop",
  "homepageUrl": null,
  "license": "MIT License",
  "language": "TypeScript",
  "starCount": 12,
  "sourceUpdatedAt": "2026-06-18T09:32:21Z"
}

💳 Pricing

This Actor uses pay-per-event pricing. You are charged for each successfully scraped Glama MCP server saved to the dataset.

You are not charged for empty searches, unknown server IDs, duplicate rows skipped inside a run, startup, or failed lookups that do not save a server row.

⚠️ Limits and notes

• This Actor is Glama-only. It does not aggregate Smithery, MCP.so, PulseMCP, or other MCP directories.
• It returns public registry and public repository facts. It does not test private MCP servers, call MCP tools, or validate user credentials.
• Repository language, stars, homepage, and update timestamps are included when the public source repository exposes them.
• Readiness fields are mechanical and source-backed. They are not AI recommendations.
• tools, resources, and prompts stay empty when Glama does not expose source-backed records for a server.

❓ FAQ

Can I scrape the whole Glama MCP registry?

Yes. Use Collect catalog and raise Maximum servers to the number of rows you want. For a first run, start smaller so you can inspect the output shape.

Can I search for MCP servers by topic?

Yes. Use Search queries with phrases such as github, database, browser, or slack. The Actor saves matching Glama MCP server rows.

Does this require a Glama API key?

No. The Actor uses public Glama MCP registry data and public repository metadata. You do not need to provide a Glama account, API key, cookies, or GitHub token.

Does it run as an MCP server?

No. This is an Apify Actor that exports Glama MCP registry data to a dataset. It does not expose live MCP tools or call MCP servers.

📝 Changelog

• 0.0: Initial release.

🆘 Support

For issues, questions, or feature requests, file a ticket and I'll fix or implement it in less than 24h 🫡

🔗 Other actors

• ChatGPT Marketplace Scraper ↗ - Export public custom GPT listings, creators, categories, capabilities, and usage facts.
• Chrome Extensions Scraper ↗ - Collect Chrome Web Store extension metadata, manifests, permissions, media, and publisher facts.
• Website URL Crawler ↗ - Crawl public websites and export rendered links, sitemap URLs, depth, and status data.
• Schema Markup Validator ↗ - Audit public pages for JSON-LD, Microdata, RDFa, meta tags, and rich-result readiness.
• Sitemap Sniffer ↗ - Find public sitemap files, sitemap indexes, URL counts, and sitemap metadata.

Made with ❤️ by Maxime Dupré