Glama MCP Registry Scraper

🔎 Glama MCP Registry Scraper for Server Data

This Actor is for developers, AI tooling teams, registry builders, and product researchers who need public Glama MCP server data. It saves one dataset row for each first eligible server match, with identity, descriptions, Glama attributes, MCP capability data, environment settings, repository facts, and mechanical readiness fields. Use it to search or browse the registry and export records for analysis and repeat lookup.

🎯 Use cases

• Find public Glama registry listings with Glama MCP Registry Scraper.
• Search for servers by topic with Glama MCP Server Search.
• Browse catalog entries with MCP Server Catalog.
• Look up an exact server by ID with MCP Server Details.
• Export source fields for a server with MCP Server Metadata.
• Explore a public directory of MCP servers with MCP Server Directory.

📦 Returned data

Each saved row is one Glama MCP server. Rows keep the source-native name, namespace, slug, description, attributes, repository links, license, language, star count, and source update time when Glama exposes them. They also include environment variables, the source environment schema, mechanical readiness facts, and source-backed tools, resources, and prompts when those records are available.

If a server matches more than once from submitted search terms or IDs, the Actor saves the first eligible match and ignores later matches. It does not combine later matches into the saved row.

🚀 Run the Actor

Set target to catalog to browse the Glama catalog, search to find servers by text, or serverDetails to look up exact Glama server IDs. Add search terms to searchQueries, exact IDs to serverIds, and filters when you use Catalog or Search. Values in sections for other targets are ignored.

Start with the prefilled github search query and the default Maximum servers value of 144 for a first run. Then open the dataset in the run results.

🧾 Input

Fill the target and the fields that belong to it. For Catalog and Search, filters narrow the matching servers. For Server details, use one or more exact Glama server IDs.

Field	Type	What it does
target	string	Chooses Catalog, Search, or Server details.
searchQueries	array of strings	Searches server names, descriptions, namespaces, and Glama attributes. Use it with Search.
serverIds	array of strings	Lists exact Glama server IDs to look up. Use it with Server details.
attributeFilters	array of strings	For Catalog and Search, keeps servers with all listed Glama attributes.
languageFilters	array of strings	For Catalog and Search, keeps servers whose repository uses one of the listed primary languages.
maxItems	integer	For Catalog and Search, stops after this many saved servers. The default is 144. Leave it empty to return all available results until the source is exhausted.

Default input from a successful current-beta run:

{
  "target": "search",
  "searchQueries": [
    "github"
  ],
  "maxItems": 144
}

📤 Output

The Actor returns one dataset row for each saved Glama MCP server. All targets use the same row shape. Nullable fields stay null when the source does not provide them.

Field	Type	What it does
serverId	string	Stable Glama server ID for audit and repeat lookup.
name	string	Source-native MCP server name.
namespace	string or null	Source namespace or owner label.
slug	string or null	Source Glama slug.
description	string or null	Source server description.
attributes	array of strings	Source-native Glama attribute labels in source order.
readiness	object	Mechanical readiness facts from visible source metadata.
readiness.hosting	string or null	Source-backed hosting classification.
readiness.score	integer or null	Mechanical score from visible source facts when safely derivable.
readiness.reasons	array of strings	Source facts used for the readiness score.
tools	array of objects	Source-backed MCP tool records.
tools[].name	string	Source-provided tool name.
tools[].description	string or null	Source-provided tool summary.
tools[].inputSchema	object or null	Source-provided JSON Schema for tool input.
resources	array of objects	Source-backed MCP resource records.
resources[].name	string or null	Source-provided resource name.
resources[].description	string or null	Source-provided resource summary.
resources[].uri	string or null	Source-provided resource URI.
prompts	array of objects	Source-backed MCP prompt records.
prompts[].name	string	Source-provided prompt name.
prompts[].description	string or null	Source-provided prompt summary.
prompts[].arguments	array of objects	Source-provided prompt arguments in source order.
prompts[].arguments[].name	string	Source-provided argument name.
prompts[].arguments[].description	string or null	Source-provided argument summary.
prompts[].arguments[].required	boolean or null	Whether the source marks the argument as required.
installCommand	string or null	Source-provided install or configuration command.
environment	object	Environment variables and source environment schema.
environment.requiredVariables	array of strings	Variables the source marks as required.
environment.optionalVariables	array of strings	Variables the source does not mark as required.
environment.schema	object or null	Source-provided environment schema.
sourceRepositoryUrl	string or null	Public source repository URL.
homepageUrl	string or null	Public homepage or project URL.
license	string or null	Source-provided license label.
language	string or null	Source-provided primary repository language.
starCount	integer or null	Source-provided repository star count.
sourceUpdatedAt	string or null	Source-native update time exposed by Glama.

Full row from a successful current-beta run:

{
  "serverId": "e2fk6ip2ek",
  "name": "Google Documents MCP",
  "namespace": "0PValencia",
  "slug": "google-workspace-mcp",
  "description": "An MCP server that enables AI assistants to manage Google Docs through natural language, including listing, reading, creating documents, and appending text. It features a simple browser-based OAuth login and persistent sessions.",
  "attributes": [
    "hosting:local-only"
  ],
  "readiness": {
    "hosting": "local-only",
    "score": 60,
    "reasons": [
      "local-only hosting"
    ]
  },
  "tools": [],
  "resources": [],
  "prompts": [],
  "installCommand": null,
  "environment": {
    "requiredVariables": [],
    "optionalVariables": [],
    "schema": {
      "properties": {},
      "type": "object",
      "required": []
    }
  },
  "sourceRepositoryUrl": "https://github.com/0PValencia/google-workspace-mcp",
  "homepageUrl": null,
  "license": "MIT License",
  "language": null,
  "starCount": null,
  "sourceUpdatedAt": null
}

💳 Pricing

This Actor uses pay-per-event pricing. One event covers each successfully saved Glama MCP server. The current event price is shown in the Actor Store.

🔌 Integrations

Use the dataset through the Apify API, download JSON, CSV, or Excel, schedule runs, or connect the data to other tools with Apify integrations.

⚠️ Limits and notes

• This Actor collects Glama data only. It does not combine Smithery, MCP.so, PulseMCP, the official MCP Registry, or other directories into the dataset.
• It returns public registry and public repository facts. It does not access private data, connect to listed MCP servers, call MCP tools, or validate user credentials.
• A missing source fact stays unavailable. For example, repository language, stars, homepage, update time, and capability records can be null or empty when Glama does not expose them.
• Readiness fields are mechanical and source-backed. They are not AI recommendations or security checks.

❓ FAQ

What if a server matches more than one search query?

The first eligible match is saved. Later matches for the same server are ignored, so one row does not collect evidence from later matches.

Can I look up a server when I know its Glama ID?

Yes. Set target to serverDetails and add one or more exact IDs to serverIds.

What does an empty Maximum servers field do?

It returns all available results until the source is exhausted. This work limit applies to Catalog and Search.

Why are tools, resources, or prompts empty?

Those arrays stay empty when Glama does not expose source-backed records for the server. The Actor does not invent capability data.

Do I need a Glama account or API key?

No. The Actor reads public Glama registry and public repository data, so you do not need a Glama account, API key, cookies, or private workspace access.

Does this Actor call the listed MCP servers?

No. It exports registry data and does not connect to or execute the listed servers or their tools.

Can I collect Smithery or MCP.so in the same dataset?

No. This Actor is Glama-only. Use a registry-specific Actor for another directory.

What happens when repository facts are missing?

The related fields stay null instead of being guessed. The row keeps the source data that Glama provides.

📝 Changelog

• 0.1: Add unified target selection for catalog, search, and exact server lookups, allow uncapped catalog and search exports, and improve recovery from interruptions and transient source failures.
• 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 🫡

🔗 Related Actors

• Smithery MCP Scraper — Collect public Smithery MCP servers by catalog, keyword, or exact server ID.
• Smithery MCP Registry Scraper — Compare public Smithery registry records with this Glama export.
• MCP.so Server Directory Scraper — Collect public MCP.so directory listings and repository data.
• MCP Registry API — Server Metadata Export — Export server metadata from the official MCP Registry.
• MCP Server Registry and Security Scorer — Review registry profiles with package and risk signals.

Made with ❤️ by Maxime Dupré