MCP Server
The MCP Server is an AI service proxy built on Cloudflare Workers that routes requests to multiple AI providers through a unified API.
Core Capabilities
Multi-provider AI integration: Proxy requests to Anthropic Claude and OpenAI through a single endpoint
Edge deployment: Global, low-latency responses via Cloudflare Workers
Type-safe: Implemented in TypeScript with the Hono framework
Cross-origin support: Built-in CORS for web application integration
API Endpoints
GET /health— Check server status and configurationGET /api/provider— Retrieve current AI provider and model configurationPOST /api/mcp— Send AI service requests with context, query, and options
Available Tools (accessible via /api/mcp)
api-client— Handle interactions with external APIsdata-processor— Process and transform data through the AI pipelinefile-handler— Manage file-related operationsmy-tool— Flexible custom message processingexample_tool— General-purpose message processing
Provides global edge deployment for the MCP server, offering low-latency proxying of AI requests through Cloudflare's distributed network
Enables routing requests to OpenAI's models through the MCP server, providing access to OpenAI's AI capabilities via a unified proxy interface
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@MCP Serverask Claude to summarize this document about AI ethics"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP Server
A modern AI service proxy built with Cloudflare Workers and Hono framework, supporting multiple AI providers including Anthropic Claude and OpenAI.
Features
Multi-provider AI service integration (Anthropic Claude, OpenAI)
Built on Cloudflare Workers for global edge deployment
Fast and efficient request handling with Hono framework
Type-safe implementation with TypeScript
CORS support for cross-origin requests
Health check and provider info endpoints
Related MCP server: Remote MCP Server Authless
Prerequisites
Node.js (LTS version recommended)
npm or pnpm package manager
Cloudflare account for deployment
API keys for supported AI providers
Installation
Clone the repository
Install dependencies:
pnpm installEnvironment Setup
Copy the example environment file:
cp .env.example .envConfigure your environment variables in
.envwith your API keys and preferences
Development
Start the development server:
pnpm run devThe server will start in development mode with hot reloading enabled.
Deployment
Deploy to Cloudflare Workers:
pnpm run deployAPI Endpoints
Health Check
GET /healthReturns server status and configuration
Provider Info
GET /api/providerReturns current AI provider and model configuration
MCP API
POST /api/mcpMain endpoint for AI service requests
Accepts JSON payload with context, query, and options
Project Structure
├── src/
│ ├── controllers/ # Request handlers
│ ├── models/ # Type definitions
│ ├── services/ # AI service implementations
│ └── index.ts # Main application entry
├── public/ # Static assets
└── wrangler.jsonc # Cloudflare Workers configurationLicense
MIT
Available Tools
5 toolsapi-clientD
ApiClient tool description
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Message to process |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description bears full responsibility for disclosing behavioral traits. It fails to mention any side effects, authorization needs, rate limits, or return characteristics, offering only the generic name.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is short but wastes its single sentence on a tautology. It is under-specified rather than concise, as it omits essential information that a longer description would provide.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's generic name and the presence of siblings, the description is wholly incomplete. It does not mention the API being communicated with, the expected output, or any special behavior, rendering it insufficient for an agent to use correctly.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the parameter 'message' is already documented in the input schema. The description adds no additional meaning or constraints beyond what the schema provides.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description 'ApiClient tool description' is a tautology that merely restates the tool name without specifying any verb, resource, or action. It does not distinguish this tool from siblings like 'data-processor' or 'file-handler'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool vs. alternatives. The description lacks any context, prerequisites, or examples, leaving the agent without any decision-making support.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
data-processorD
DataProcessor tool description
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Message to process |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations, the description carries the full burden of behavioral disclosure. It tells nothing about side effects, permissions, restrictions, or return behavior, leaving the agent completely in the dark about what the tool does.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely short (4 words) but wastes the opportunity to convey any useful information. Conciseness at the expense of substance is not valued.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given low complexity (1 optional param, no annotations, no output schema), a minimal description should at least clarify the nature of processing. This description is completely inadequate.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100% (the 'message' parameter has a description), so baseline is 3. The tool description adds no further meaning beyond the schema, but does not harm.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description 'DataProcessor tool description' is a tautology that merely restates the tool name without specifying any action or resource. It fails to convey what data processing actually entails.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus its siblings (api-client, example_tool, file-handler, my-tool). There is no mention of context, prerequisites, or alternatives.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
example_toolC
An example tool that processes messages
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Message to process |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations provided and the description lacks any behavioral traits such as side effects, permissions, or read-only/mutating nature.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
While short, the description is under-specified and does not earn its place with substantive information.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the simple schema and no output schema or annotations, the description fails to provide complete context, e.g., what the output is or expected behavior.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, so the description adds no value beyond the schema definition for the only parameter 'message'.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description 'processes messages' is vague and does not specify what kind of processing or distinguish from sibling tools like api-client, data-processor, and file-handler.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance on when to use this tool versus alternatives; no exclusions or prerequisites mentioned.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
file-handlerD
FileHandler tool description
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Message to process |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description must disclose behavioral traits. It contains no information about side effects, permissions, or any behavior beyond the name, leaving the agent completely uninformed.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is very short, but it is underspecified rather than concise. It fails to convey useful information about the tool's purpose or usage, making it inefficient for an agent.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (single parameter, no output schema), the description should at least state what the tool does with the message. It does not, leaving the context completely incomplete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The schema description coverage is 100% (one parameter 'message' described as 'Message to process'). The description adds no additional meaning beyond the schema, so a baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description 'FileHandler tool description' is a tautology, merely restating the tool name without specifying any verb or resource. It does not clarify what the tool does, and it fails to distinguish it from siblings like 'api-client' or 'data-processor'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool, when not to, or how it compares to alternatives. The description is generic and offers no context for selection.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
my-toolD
MyTool tool description
| Name | Required | Description | Default |
|---|---|---|---|
| message | No | Message to process |
TDQS
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations available, the description carries full responsibility for behavioral disclosure. It merely repeats the tool name and offers no insight into side effects, permissions, or constraints.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
While the description is short (three words), it is under-specified rather than concise. It lacks front-loaded key information and fails to earn its place by adding value.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's single parameter, lack of output schema, and absent annotations, the description is grossly incomplete. It provides no operational context for agent invocation.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, and the input schema already describes the 'message' parameter. The description adds no further semantic value beyond the schema, so a baseline score of 3 is appropriate.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description 'MyTool tool description' is a tautology that restates the tool name without indicating any specific action or resource. It fails to distinguish from sibling tools like 'api-client' or 'data-processor'.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No usage context or guidance is provided. The description does not specify when to use this tool over alternatives, nor does it mention any prerequisites or exclusions.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
5 tool updates
- First observed
api-client - First observed
data-processor - First observed
example_tool - First observed
file-handler - First observed
my-tool
TDQS
Scored across 5 tools
The tools have distinct names like api-client, data-processor, and file-handler, which suggest different purposes, but descriptions are generic (e.g., 'ApiClient tool description'), leaving some ambiguity about their specific functions. Without detailed descriptions, it's unclear if there's overlap in processing or handling tasks.
Naming is inconsistent with mixed conventions: hyphen-case (api-client, data-processor, my-tool), snake_case (example_tool), and camelCase (FileHandler). This lack of a predictable pattern makes the tool set harder to navigate and could confuse agents.
With 5 tools, the count is reasonable for a general-purpose MCP server, suggesting a moderate scope. It's not too thin or heavy, though the generic nature of the tools makes it hard to assess if this number fits a specific domain perfectly.
Given the generic tool names and descriptions, it's impossible to infer a clear domain, making completeness assessment difficult. However, the set lacks obvious CRUD or lifecycle coverage, and tools like 'example_tool' and 'my-tool' seem vague, indicating significant gaps for any specific workflow.
Maintenance
Related MCP Connectors
Cloudflare Workers MCP server: ai-model-router
Cloudflare Workers MCP server: ai-gateway
AI routing, memory, guardrails, and governance. Routes across Claude, GPT, Gemini.
Cloudflare Workers MCP server: ai-guardrails
Related MCP Servers
- FlicenseNot gradedqualityDmaintenanceAn open standard server implementation that enables AI assistants to directly access APIs and services through Model Context Protocol, built using Cloudflare Workers for scalability.-
- FlicenseNot gradedqualityCmaintenanceA serverless Model Context Protocol implementation on Cloudflare Workers that enables AI tools without requiring authentication, allowing connection from AI Playground or Claude Desktop.-
- FlicenseNot gradedqualityCmaintenanceA deployable Model Context Protocol server on Cloudflare Workers that enables custom AI tools without requiring authentication, compatible with Cloudflare AI Playground and Claude Desktop.-
- FlicenseNot gradedqualityCmaintenanceA deployable MCP server on Cloudflare Workers that enables AI tools integration without authentication requirements, compatible with AI Playground and Claude Desktop.-