Glama

What is Glama?

Glama is a registry, inspector, and gateway for Model Context Protocol servers. It indexes tens of thousands of open source servers and hosted connectors, lets you test any of them in the browser, and puts a managed proxy in front of every connection so teams keep logging and access control together.

Top Features:

• Server registry: browse a searchable index of open source servers and connectors.
• Browser inspector: test any server in a sandbox without installing anything locally.
• Managed hosting: deploy servers on managed infrastructure with logging and access rules.
• Gateway proxy: route tool calls through one endpoint with credentials handled centrally.

Use Cases:

• Tool discovery: find a working server for a given data source quickly.
• Security review: check quality scores and audits before wiring a server in.
• Usage tracking: watch which tools agents call and how often they run.

Who Can Use Glama?

• AI developers: connect assistants to real data without building every integration first.
• Platform teams: keep credentials and tool permissions under central control across projects.
• Product engineers: ship features that depend on outside tools with less plumbing.

Pricing

• Starter ($9 per month): personal use with chat, custom agents, three hosted servers.
• Pro ($26 per month): adds file uploads, projects, memory, web search, ten servers.
• Business ($80 per month): custom exports, priority support, and thirty hosted servers included.

Pros and Cons

Pros:

• Huge catalog: the index covers far more servers than the official registry.
• No install testing: the inspector saves setup time when comparing similar servers.
• Central control: per tool permissions make audits easier for security minded teams.

Cons:

• Credit based billing: monthly AI credits can run out during busy build weeks.
• Technical audience: people outside engineering will struggle with the protocol concepts.
• Extra server fees: hosting beyond the plan limit adds cost per instance.

FAQs:

1) What is an MCP server?

It is a small service that gives AI clients access to outside tools and data.

2) Do I need to write code?

Developer knowledge helps, though many hosted connectors work with simple one click setup.

3) Is hosting free for open source servers?

Yes, open source servers receive free hosting and deployment on the platform.

4) Which AI clients can connect?

Claude, ChatGPT, Cursor, Zed, VS Code, and other protocol aware clients all connect.

5) Can I track tool usage?

Yes, the gateway records logs and usage analytics for every connected server.