Glama MCP. Find, connect, and test any external AI model from a single API gateway.
Gemini Works with every AI agent you already use
…and any MCP-compatible client
Just plug in your AI agents and start using Vinkius.
Glama connects your AI agent to an external directory of MCP servers. It lets your agent dynamically discover, index, and analyze attributes of external intelligence networks without manual setup.
You can query multiple standard LLM networks via the Glama API Gateway, consolidating programmatic text completion requirements into one unified call.
What your AI agents can do
Glama get gateway model details
Fetches specific details (like price or context window) for a named model proxy on the Glama gateway.
Glama get gateway models
Lists all OpenAI-compatible AI models currently supported by the Glama gateway.
Glama get hosted instances
Retrieves a list of private MCP instances assigned to your specific Glama account.
Use glama_list_mcp_servers to search the global directory for available context protocols and services.
Call glama_get_mcp_server_info with a namespace and slug to get a server's full parameters and setup instructions.
Run glama_get_gateway_models to list every OpenAI-compatible model currently available through the Glama gateway.
Execute a conversation prompt using glama_run_gateway_chat via a specific model proxy, keeping the logic outside your local environment.
Run glama_get_hosted_instances to fetch all private MCP instances assigned directly to your Glama account.
Use glama_get_mcp_attributes to list and filter semantic categorizations across the entire MCP Registry.
Ask AI about this MCP
Supported MCP Clients
Gemini Waiting for input…
Glama MCP Server: 8 Tools for AI Model Management
These tools let your agent discover, verify, and connect to external AI models and services within the Vinkius ecosystem.
019d75a6glama get gateway model details
Fetches specific details (like price or context window) for a named model proxy on the Glama gateway.
019d75a6glama get gateway models
Lists all OpenAI-compatible AI models currently supported by the Glama gateway.
019d75a6glama get hosted instances
Retrieves a list of private MCP instances assigned to your specific Glama account.
019d75a6glama get mcp attributes
Lists the semantic categories and filtering attributes mapped within the global Glama MCP Registry.
019d75a6glama get mcp server info
Extracts detailed parameters and setup instructions for one specific MCP server, given its namespace and slug.
019d75a6glama list mcp servers
Searches and lists MCP servers from the global Glama directory using loose text matching.
019d75a6glama run gateway chat
Sends a conversational prompt to an isolated model via the Glama proxy network.
019d75a6glama send telemetry
Reports usage metrics and execution data back to the Glama Telemetry backend after a tool runs.
Choose How to Get Started
Build a custom MCP for your own tools, or connect a ready-made integration from our catalog.
Build Your Own
Turn any API into an MCP. Import a spec, define Agent Skills, or deploy with MCPFusion.
- Import from OpenAPI, Swagger, or YAML specs
- Create Agent Skills with progressive disclosure
- Deploy to edge with MCPFusion framework
- Built in DLP, auth, and compliance on every call
- Real time usage dashboard and cost metering
- Publish to catalog or keep private
Make Your AI Do More
Start with Glama, then connect any of our 4,700+ other servers whenever your AI needs more. One click, no limits.
- Use this MCP plus 4,700+ others, all in one place
- Add new capabilities to your AI anytime you want
- Every connection is secured and compliant automatically
- Track usage and costs across all your servers
- Works with Claude, ChatGPT, Cursor, and more
- New servers added to the catalog every week
What you can do with this MCP connector
Glama MCP Server: The AI Infrastructure Bridge
This thing hooks up your AI agent to a whole external directory of MCP servers. Your agent can dynamically find, index, and analyze attributes from external intelligence networks without you having to set up anything manually. You can also send your agent to query multiple standard LLM networks using the Glama API Gateway, consolidating all programmatic text completion requirements into one unified call.
Discovering external MCP servers
Use glama_list_mcp_servers to search the global directory for context protocols and services. You can then call glama_get_mcp_server_info with a namespace and slug to get a server's full parameters and setup instructions. To check what kind of categories are out there, run glama_get_mcp_attributes to list and filter semantic categorizations across the entire MCP Registry.
You can also run glama_get_hosted_instances to fetch all private MCP instances assigned directly to your Glama account. To know what models are playing on the gateway, run glama_get_gateway_models to list every OpenAI-compatible model currently available through the Glama gateway. You can execute a conversation prompt using glama_run_gateway_chat via a specific model proxy, keeping the logic outside your local environment.
You'll also use glama_get_gateway_model_details to fetch specific details like a model's price or context window for a named proxy. Finally, you send usage metrics and execution data back to the Glama Telemetry backend after a tool runs using glama_send_telemetry.
How Glama MCP Works
- 1 Mount the Glama logic layer inside your Vinkius limits.
- 2 In your Glama settings UI, emit an API token. Map it to the
GLAMA_API_KEYvariable. - 3 Instruct your agent: 'Identify 3 active finance MCPs from the Glama network. Also, extract the context window sizes of Claude using the Gateway module.' The agent handles the sequential calls.
The bottom line is that Glama acts as a single, smart front-end that lets your agent talk to dozens of specialized services without you writing connection code for each one.
Who Is Glama MCP For?
The architectural DevOps Engineer who needs to prototype and connect disparate API models without dealing with dashboard clutter. It’s for the core Financial Analyst who needs to locate specific enterprise integrations. Or the Asymmetric Operations Manager who needs to map complex logic attributes across multiple services.
Prototyping and executing API models, isolating specific protocols, and avoiding manual dashboard UI noise.
Locating specific enterprise MCP integrations and mapping variables that simulate external data endpoints.
Extracting and mapping metric attributes from hosted proxies to limit execution friction during complex operations.
What Changes When You Connect
- Consolidate model discovery. Instead of jumping between multiple vendor documentation sites,
glama_list_mcp_serverssearches the entire directory, letting your agent find the right protocol instantly. - Handle complex conversations. Use
glama_run_gateway_chatto send a prompt to a specific model proxy (like Claude or Llama) without having to manage the connection parameters yourself. - Audit all available models.
glama_get_gateway_modelsgives you a complete list of every OpenAI-compatible proxy available, which is a massive time saver for initial setup. - Understand model constraints.
glama_get_gateway_model_detailsprovides granular data—like context window size or pricing—for a model before you commit to using it. - Map your infrastructure.
glama_get_mcp_attributeslets you query the global logic matrix, helping you categorize and find servers based on abstract concepts (e.g., 'finance' or 'B2B'). - Track usage automatically. Use
glama_send_telemetryafter a successful run to report usage metrics back to Glama, keeping your action matrix documented.
Real-World Use Cases
Mapping a new financial data source
A Financial Analyst needs to connect to a new B2B data source. Instead of manually checking documentation, they run glama_list_mcp_servers. The agent finds 'Finance-MCP' and then uses glama_get_mcp_server_info to pull the exact parameters needed for the integration, solving the setup bottleneck.
Benchmarking model performance
A DevOps Engineer needs to know if Llama 3 8B is faster than Claude Haiku for a specific task. They first run glama_get_gateway_models to see both are available, then use glama_get_gateway_model_details to pull their context window sizes and pricing, enabling a direct, data-driven comparison.
Building a multi-step agent workflow
An Operations Manager needs the agent to perform a multi-stage task: 1) find all CRM tools (glama_list_mcp_servers), 2) ask for their attributes (glama_get_mcp_attributes), and 3) run a test query on the best one (glama_run_gateway_chat). Glama coordinates the entire sequence, eliminating manual orchestration.
Debugging a failed external call
The agent fails on a service and the user needs to know why. They run glama_send_telemetry to report the failure event back to the Glama backend. This logs the full usage context and failure metrics, making debugging systematic.
The Tradeoffs
Copying credentials manually
The developer gets the API endpoint for the 'Salesforce-MCP' from one document, the LLM model name from another, and the proxy details from a third. They spend an hour just assembling the initial function call.
→
Let the agent do the heavy lifting. First, use glama_list_mcp_servers to find the server slug. Then, use glama_get_mcp_server_info to pull all required parameters in one call. This keeps the data flow inside the Vinkius environment.
Assuming model compatibility
Trying to send a message to a model because it sounds similar to a model you already use, only to find out the context window size is wrong or the pricing model doesn't match the task requirements.
→
Always check the specs first. Run glama_get_gateway_models to list all options, then use glama_get_gateway_model_details to confirm the context window and cost structure before running any chat (glama_run_gateway_chat).
Sequential, unvalidated calls
Running glama_list_mcp_servers and then manually reading the output to pick a slug, only to find out that the slug is deprecated or invalid for the current task.
→
Always validate the context. Use glama_get_mcp_attributes first to filter the list of servers by the required category (e.g., 'finance'). This reduces the pool of options and limits the risk of invalid selection.
When It Fits, When It Doesn't
Use Glama if your job requires connecting to multiple, disparate external AI services (CRM, Finance, etc.) and you need to discover their specific parameters dynamically. It’s perfect for building multi-step, complex agent workflows. Don't use it if you only need to call a single, known API endpoint—a direct SDK integration is faster. If you just need to list models, glama_get_gateway_models handles that. If you need to check a specific server's setup, use glama_get_mcp_server_info. If your goal is simply text completion with one model, glama_run_gateway_chat works, but you'll miss the full discovery power of the other tools.
Independent Platform Disclaimer: Vinkius is an independent platform and is not affiliated with, endorsed by, sponsored by, verified by, or otherwise authorized by Glama. All third-party trademarks, logos, and brand names are the property of their respective owners. Their use on this website is strictly for informational purposes to identify service compatibility and interoperability.
VINKIUS INFRASTRUCTURE
Cloud Hosted
Managed infra
V8 Isolated
Sandboxed per request
Zero-Trust Proxy
No stored credentials
DLP Enforced
Policy on every call
GDPR Compliant
EU data residency
Token Compression
~60% cost reduction
Works with Claude, ChatGPT, Cursor, and more
The Model Context Protocol standardizes how applications expose capabilities to LLMs. Instead of operating in isolation, your AI gains direct access to external platforms, live data, and real-world actions through secure, standardized connections.
This server provides 8 capabilities that interface natively with Claude, ChatGPT, Cursor, and any MCP client. No middleware. No custom integration required.
Available Capabilities
Finding the right AI model connection shouldn't require a dozen tabs and five different API keys.
Today, if you need to connect your agent to a new external service—say, a niche CRM or a specific financial data feed—you're stuck. You have to open the vendor's documentation, find the specific MCP server slug, and then manually look up its required parameters (auth headers, context limits, etc.). It’s a painful copy-paste cycle that kills deep work.
With Glama, you just let your agent run `glama_list_mcp_servers`. It searches the whole directory for you. Then, if you need details, `glama_get_mcp_server_info` pulls the exact specs. You get a single, clean source of truth, eliminating manual research and credential management.
glama_run_gateway_chat: Chatting with any model, no matter where it lives
Before Glama, if your agent needed to talk to a smaller, specialized LLM (like Llama 3 8B), you had to write a whole new block of code, managing the endpoint, the proxy keys, and the connection logic every time. It was repetitive boilerplate.
Now, you just use `glama_run_gateway_chat`. You specify the model, and Glama handles the proxy network connection. It’s instant, reliable, and keeps your core logic clean. You just tell it what you want it to say.
Common Questions About Glama MCP
How do I find out what MCP servers are available using glama_list_mcp_servers? +
Run glama_list_mcp_servers to search the global directory. This tool lets your agent find available context protocols by name or category without you needing to know the exact server slug beforehand.
What is the difference between glama_get_mcp_server_info and glama_get_mcp_attributes? +
Use glama_get_mcp_server_info when you know the server's name (slug) and need its specific setup parameters. Use glama_get_mcp_attributes when you want to browse and filter all servers by a general category, like 'finance' or 'B2B'.
Can I check the context window size of a model using glama_get_gateway_model_details? +
Yes. glama_get_gateway_model_details fetches granular attributes for a specific model proxy. This allows you to compare things like context window size and pricing before you actually run the chat.
How do I run a chat prompt on a model using glama_run_gateway_chat? +
You trigger glama_run_gateway_chat and provide the prompt and the model name. Glama manages the connection through its proxy network, giving you the response without local setup hassle.
Where should I report my usage metrics using glama_send_telemetry? +
You call glama_send_telemetry after your agent has completed a task using an external server. This reports the usage event back to the Glama backend, helping you track performance and usage.
How do I fetch my private instances using glama_get_hosted_instances? +
It fetches all private MCP instances assigned to your account. You use this tool when you need to list the specific, non-public servers you've set up within the Glama ecosystem.
What is the purpose of glama_get_gateway_models? +
This tool audits the full list of AI models supported by the OpenAI-compatible gateway. You run this when you need to know which models are available for chat or proxying.
How do I check for compatible external services using glama_get_mcp_attributes? +
It lists the filtering attributes and semantic categorizations mapped in the Glama MCP Registry. This lets you quickly filter the directory for specific types of logic, like finance or CRM.
Can I test alternative AI models entirely within the terminal using the Glama integration? +
Yes. Tools like glama_get_gateway_models list available OpenAI-compatible proxies, and glama_run_gateway_chat allows your Vinkius agent to run text completions outside itself natively.
Does the Glama server provide telemetry data back to the registry? +
Yes. Active MCP usage events can be logged seamlessly applying the glama_send_telemetry tool in specific sequences to inform publishers about proxy executions.
Are private hosted instances queryable? +
Yes. By executing glama_get_hosted_instances, your agent limits queries exclusively to private proxies explicitly belonging to your linked environment.
Use it with your favorite AI tools
Connect this server to Cursor, Claude, VS Code, and more.
More in this category
Chainlit
Empower your AI agents to audit chat threads, analyze model steps, and track LLM observability metrics securely.
Tray.io
Equip your AI agent to orchestrate automations, track active workflows, and monitor data execution flows across Tray.io natively.
Haystack (deepset Cloud)
Build and manage AI-powered search and RAG pipelines via deepset Cloud — search documents, run pipelines, and manage workspaces.
You might also like
Make.com Webhook Trigger
This MCP does exactly one thing: it sends JSON payloads to Make.com Webhooks. That's its only function. Incredible for connecting AI agents to thousands of visual automation workflows instantly.
Sobot
Leading AI customer support and ticketing platform in China — manage tickets, agents, and knowledge via AI.
Everbridge Critical Management
Equip your AI agent to manage critical notifications, track incidents, and monitor contacts via the Everbridge API.