Mcp
Experiments MCP
Model Context Protocol reference for experiments.do - Test and iterate functions, workflows, and agents
Experiments MCP
Test and iterate functions, workflows, and agents
Overview
The Model Context Protocol (MCP) provides AI models with direct access to experiments.do through a standardized interface.
Installation
pnpm add @modelcontextprotocol/sdkConfiguration
Add to your MCP server configuration:
{
"mcpServers": {
"experiments": {
"command": "npx",
"args": ["-y", "@dotdo/mcp-server"],
"env": {
"DO_API_KEY": "your-api-key"
}
}
}
}Tools
experiments/invoke
Main tool for experiments.do operations.
{
"name": "experiments/invoke",
"description": "Test and iterate functions, workflows, and agents",
"inputSchema": {
"type": "object",
"properties": {
"operation": {
"type": "string",
"description": "Operation to perform"
},
"parameters": {
"type": "object",
"description": "Operation parameters"
}
},
"required": ["operation"]
}
}Usage in AI Models
Claude Desktop
// ~/Library/Application Support/Claude/claude_desktop_config.json
{
"mcpServers": {
"experiments": {
"command": "npx",
"args": ["-y", "@dotdo/mcp-server", "--tool=experiments"],
"env": {
"DO_API_KEY": "undefined"
}
}
}
}OpenAI GPTs
# Custom GPT configuration
tools:
- type: mcp
server: experiments
operations:
- invoke
- query
- executeCustom Integration
import { Client } from '@modelcontextprotocol/sdk/client/index.js'
import { StdioClientTransport } from '@modelcontextprotocol/sdk/client/stdio.js'
const transport = new StdioClientTransport({
command: 'npx',
args: ['-y', '@dotdo/mcp-server', '--tool=experiments'],
})
const client = new Client(
{
name: 'experiments-client',
version: '1.0.0',
},
{
capabilities: {},
}
)
await client.connect(transport)
// Call tool
const result = await client.callTool({
name: 'experiments/invoke',
arguments: {
operation: 'experiments',
parameters: {},
},
})Tool Definitions
Available Tools
{
"tools": [
{
"name": "experiments/invoke",
"description": "Invoke experiments.do",
"inputSchema": {
/* ... */
}
},
{
"name": "experiments/query",
"description": "Query experiments.do resources",
"inputSchema": {
/* ... */
}
},
{
"name": "experiments/status",
"description": "Check experiments.do status",
"inputSchema": {
/* ... */
}
}
]
}Resources
Available Resources
{
"resources": [
{
"uri": "experiments://config",
"name": "Experiments Configuration",
"mimeType": "application/json"
},
{
"uri": "experiments://docs",
"name": "Experiments Documentation",
"mimeType": "text/markdown"
}
]
}Prompts
Pre-configured Prompts
{
"prompts": [
{
"name": "experiments-quick-start",
"description": "Quick start guide for experiments.do",
"arguments": []
},
{
"name": "experiments-best-practices",
"description": "Best practices for experiments.do",
"arguments": []
}
]
}Examples
Basic Usage
// AI model calls tool via MCP
mcp call experiments/runWith Parameters
// Call with parameters
await mcp.callTool('experiments/invoke', {
operation: 'process',
parameters: {
// Operation-specific parameters
},
options: {
timeout: 30000,
},
})Error Handling
try {
const result = await mcp.callTool('experiments/invoke', {
operation: 'process',
})
return result
} catch (error) {
if (error.code === 'TOOL_NOT_FOUND') {
console.error('Experiments tool not available')
} else {
throw error
}
}AI Integration Patterns
Agentic Workflows
// AI agent uses experiments.do in workflow
const workflow = {
steps: [
{
tool: 'experiments/invoke',
operation: 'analyze',
input: 'user-data',
},
{
tool: 'experiments/process',
operation: 'transform',
input: 'analysis-result',
},
],
}Chain of Thought
AI models can reason about experiments.do operations:
User: "I need to process this data"
AI: "I'll use the experiments tool to:
1. Validate the data format
2. Process it through experiments.do
3. Return the results
Let me start..."
[Calls: mcp call experiments/run]Server Implementation
Custom MCP Server
import { Server } from '@modelcontextprotocol/sdk/server/index.js'
import { StdioServerTransport } from '@modelcontextprotocol/sdk/server/stdio.js'
const server = new Server(
{
name: 'experiments-server',
version: '1.0.0',
},
{
capabilities: {
tools: {},
resources: {},
prompts: {},
},
}
)
// Register tool
server.setRequestHandler('tools/call', async (request) => {
if (request.params.name === 'experiments/invoke') {
// Handle experiments.do operation
return {
content: [
{
type: 'text',
text: JSON.stringify(result),
},
],
}
}
})
const transport = new StdioServerTransport()
await server.connect(transport)Best Practices
- Tool Design - Keep tools focused and single-purpose
- Error Messages - Provide clear, actionable errors
- Documentation - Include examples in tool descriptions
- Rate Limiting - Implement appropriate limits
- Security - Validate all inputs from AI models
- Monitoring - Track tool usage and errors