A Python implementation of the Model Context Protocol (MCP) with Enhanced Tool Definition Interface (ETDI) security extensions that seamlessly integrates with existing MCP infrastructure.
This SDK provides a secure implementation of MCP with OAuth 2.0-based security enhancements to prevent Tool Poisoning and Rug Pull attacks. ETDI adds cryptographic verification, immutable versioned definitions, and explicit permission management to the MCP ecosystem while maintaining full compatibility with existing MCP servers and clients.
ETDI is designed for zero-friction adoption with existing MCP infrastructure:
- Existing MCP servers work unchanged - ETDI clients can discover and use any MCP server
- Existing MCP clients work unchanged - ETDI servers are fully MCP-compatible
- Gradual migration path - Add security incrementally without breaking existing workflows
- Optional security - ETDI features are opt-in, not mandatory
# Existing FastMCP server becomes ETDI-secured with decorator from mcp.server.fastmcp import FastMCP app = FastMCP("My Server") # Standard tool (no security) @app.tool() def standard_tool(data: str) -> str: return f"Processed: {data}" # ETDI-secured tool with OAuth + Request Signing @app.tool( etdi=True, etdi_permissions=["data:read", "data:write"], etdi_oauth_scopes=["tools:execute"], etdi_require_request_signing=True ) def secure_tool(sensitive_data: str) -> str: return f"Securely processed: {sensitive_data}"
# ETDI client discovers ALL MCP servers (ETDI and non-ETDI) from mcp.etdi.client import ETDIClient client = ETDIClient(config) await client.connect_to_server(["python", "-m", "any_mcp_server"], "server-name") tools = await client.discover_tools() # Works with any MCP server!
- Client/Server Architecture: Full MCP client and server implementations
- Tool Management: Register, discover, and invoke tools
- Resource Access: Secure access to external resources
- Prompt Templates: Reusable prompt templates for LLM interactions
- 🔄 Full MCP Compatibility: Works with any existing MCP server or client
- OAuth 2.0 Integration: Support for Auth0, Okta, Azure AD, and custom providers
- Tool Verification: Cryptographic verification of tool authenticity
- Permission Management: Fine-grained permission control with OAuth scopes
- Version Control: Automatic detection of tool changes requiring re-approval
- Approval Management: Encrypted storage of user tool approvals
- Request Signing: RSA/ECDSA cryptographic signing for enhanced security
- Security Inspector Tools: Built-in tools for security analysis and debugging
- Tool Poisoning Prevention: Cryptographic verification prevents malicious tool impersonation
- Rug Pull Protection: Version and permission change detection prevents unauthorized modifications
- Multiple Security Levels: Basic, Enhanced, and Strict security modes
- Audit Logging: Comprehensive security event logging
- Call Stack Verification: Prevents unauthorized nested tool calls
- 🛡️ Non-Breaking Security: Security features don't break existing MCP workflows
mcp dev server.py
The Model Context Protocol (MCP) lets you build servers that expose data and functionality to LLM applications in a secure, standardized way. Think of it like a web API, but specifically designed for LLM interactions. MCP servers can:
- Expose data through Resources (think of these sort of like GET endpoints; they are used to load information into the LLM's context)
- Provide functionality through Tools (sort of like POST endpoints; they are used to execute code or otherwise produce a side effect)
- Define interaction patterns through Prompts (reusable templates for LLM interactions)
- And more!
The FastMCP server is your core interface to the MCP protocol. It handles connection management, protocol compliance, and message routing:
# Add lifespan support for startup/shutdown with strong typing from contextlib import asynccontextmanager from collections.abc import AsyncIterator from dataclasses import dataclass from fake_database import Database # Replace with your actual DB type from mcp.server.fastmcp import FastMCP # Create a named server mcp = FastMCP("My App") # Specify dependencies for deployment and development mcp = FastMCP("My App", dependencies=["pandas", "numpy"]) @dataclass class AppContext: db: Database @asynccontextmanager async def app_lifespan(server: FastMCP) -> AsyncIterator[AppContext]: """Manage application lifecycle with type-safe context""" # Initialize on startup db = await Database.connect() try: yield AppContext(db=db) finally: # Cleanup on shutdown await db.disconnect() # Pass lifespan to server mcp = FastMCP("My App", lifespan=app_lifespan) # Access type-safe lifespan context in tools @mcp.tool() def query_db() -> str: """Tool that uses initialized resources""" ctx = mcp.get_context() db = ctx.request_context.lifespan_context["db"] return db.query()
Resources are how you expose data to LLMs. They're similar to GET endpoints in a REST API - they provide data but shouldn't perform significant computation or have side effects:
import asyncio from mcp.etdi import ETDIClient, OAuthConfig, SecurityLevel async def main(): # Configure OAuth provider oauth_config = OAuthConfig( provider="auth0", client_id="your-client-id", client_secret="your-client-secret", domain="your-domain.auth0.com", audience="https://your-api.example.com", scopes=["read:tools", "execute:tools"] ) # Initialize ETDI client async with ETDIClient({ "security_level": SecurityLevel.ENHANCED, "oauth_config": oauth_config.to_dict(), "allow_non_etdi_tools": True, "show_unverified_tools": False }) as client: # Connect to MCP servers await client.connect_to_server(["python", "-m", "my_server"], "my-server") # Discover and verify tools tools = await client.discover_tools() for tool in tools: if tool.verification_status.value == "verified": # Approve tool for usage await client.approve_tool(tool) # Invoke tool result = await client.invoke_tool(tool.id, {"param": "value"}) print(f"Result: {result}") asyncio.run(main())
import asyncio from mcp.etdi.server import ETDISecureServer from mcp.etdi import OAuthConfig async def main(): # Configure OAuth oauth_configs = [ OAuthConfig( provider="auth0", client_id="your-client-id", client_secret="your-client-secret", domain="your-domain.auth0.com", audience="https://your-api.example.com", scopes=["read:tools", "execute:tools"] ) ] # Create secure server server = ETDISecureServer(oauth_configs) # Register secure tool @server.secure_tool(permissions=["read:data", "write:data"]) async def secure_calculator(operation: str, a: float, b: float) -> float: """A secure calculator with OAuth protection""" if operation == "add": return a + b elif operation == "multiply": return a * b else: raise ValueError(f"Unknown operation: {operation}") await server.initialize() print("Secure server running with OAuth protection") asyncio.run(main())
from mcp.etdi import OAuthConfig auth0_config = OAuthConfig( provider="auth0", client_id="your-auth0-client-id", client_secret="your-auth0-client-secret", domain="your-domain.auth0.com", audience="https://your-api.example.com", scopes=["read:tools", "execute:tools"] )
okta_config = OAuthConfig( provider="okta", client_id="your-okta-client-id", client_secret="your-okta-client-secret", domain="your-domain.okta.com", scopes=["etdi.tools.read", "etdi.tools.execute"] )
from mcp.server.fastmcp import FastMCP, Context mcp = FastMCP("My App") @mcp.tool() async def long_task(files: list[str], ctx: Context) -> str: """Process multiple files with progress tracking""" for i, file in enumerate(files): ctx.info(f"Processing {file}") await ctx.report_progress(i, len(files)) data, mime_type = await ctx.read_resource(f"file://{file}") return "Processing complete"
Authentication can be used by servers that want to expose tools accessing protected resources.
mcp.server.auth
implements an OAuth 2.0 server interface, which servers can use by providing an implementation of the OAuthAuthorizationServerProvider
protocol.
from mcp import FastMCP from mcp.server.auth.provider import OAuthAuthorizationServerProvider from mcp.server.auth.settings import ( AuthSettings, ClientRegistrationOptions, RevocationOptions, ) class MyOAuthServerProvider(OAuthAuthorizationServerProvider): # See an example on how to implement at `examples/servers/simple-auth` ... mcp = FastMCP( "My App", auth_server_provider=MyOAuthServerProvider(), auth=AuthSettings( issuer_url="https://myapp.com", revocation_options=RevocationOptions( enabled=True, ), client_registration_options=ClientRegistrationOptions( enabled=True, valid_scopes=["myscope", "myotherscope"], default_scopes=["myscope"], ), required_scopes=["myscope"], ), )
See OAuthAuthorizationServerProvider for more details.
The fastest way to test and debug your server is with the MCP Inspector:
mcp dev server.py # Add dependencies mcp dev server.py --with pandas --with numpy # Mount local code mcp dev server.py --with-editable .
Once your server is ready, install it in Claude Desktop:
mcp install server.py # Custom name mcp install server.py --name "My Analytics Server" # Environment variables mcp install server.py -v API_KEY=abc123 -v DB_URL=postgres://... mcp install server.py -f .env
For advanced scenarios like custom deployments:
from mcp.etdi.inspector import SecurityAnalyzer analyzer = SecurityAnalyzer() # Analyze tool security result = await analyzer.analyze_tool(tool_definition) print(f"Security Score: {result.security_score}") print(f"Vulnerabilities: {result.vulnerabilities}")
from mcp.etdi.inspector import TokenDebugger debugger = TokenDebugger() # Debug JWT tokens debug_info = await debugger.debug_token(jwt_token) print(f"Token valid: {debug_info.valid}") print(f"Claims: {debug_info.claims}") print(f"Issues: {debug_info.issues}")
from mcp.etdi.inspector import OAuthValidator validator = OAuthValidator() # Validate OAuth configuration result = await validator.validate_provider("auth0", oauth_config) print(f"Configuration valid: {result.configuration_valid}") print(f"Provider reachable: {result.is_reachable}")
ETDI provides command-line tools for configuration and debugging:
# Initialize ETDI configuration python -m mcp.etdi.cli init --provider auth0 # Validate OAuth configuration python -m mcp.etdi.cli validate-oauth --config etdi-config.json # Debug JWT tokens python -m mcp.etdi.cli debug-token --token "eyJ..." # Analyze tool security python -m mcp.etdi.cli analyze-tool --tool-id "my-tool"
- Simple cryptographic verification
- No OAuth requirements
- Suitable for development and testing
- OAuth 2.0 token verification
- Permission-based access control
- Tool change detection
- Suitable for production use
- Full OAuth enforcement
- Request signing required
- No unverified tools allowed
- Maximum security for sensitive environments
- ETDIClient: Main client interface with security verification
- ETDIVerifier: OAuth token verification and change detection
- ApprovalManager: Encrypted storage of user approvals
- SecureSession: Enhanced MCP client session with security
- ETDISecureServer: OAuth-protected MCP server
- SecurityMiddleware: Security middleware for tool protection
- TokenManager: OAuth token lifecycle management
- ToolProvider: Secure tool registration and management
- Auth0Provider: Auth0 integration with JWKS validation
- OktaProvider: Okta integration with custom scopes
- AzureADProvider: Azure AD integration with tenant support
- OAuthManager: Multi-provider management and failover
- SecurityAnalyzer: Tool security analysis and scoring
- TokenDebugger: JWT token debugging and validation
- OAuthValidator: OAuth configuration validation
- CallStackVerifier: Call stack verification and analysis
ETDI supports cryptographic request signing with RSA-SHA256 signatures embedded directly in MCP protocol messages:
# main.py import contextlib from fastapi import FastAPI from mcp.echo import echo from mcp.math import math # Create a combined lifespan to manage both session managers @contextlib.asynccontextmanager async def lifespan(app: FastAPI): async with contextlib.AsyncExitStack() as stack: await stack.enter_async_context(echo.mcp.session_manager.run()) await stack.enter_async_context(math.mcp.session_manager.run()) yield app = FastAPI(lifespan=lifespan) app.mount("/echo", echo.mcp.streamable_http_app()) app.mount("/math", math.mcp.streamable_http_app())
For low level server with Streamable HTTP implementations, see:
- Stateful server:
examples/servers/simple-streamablehttp/
- Stateless server:
examples/servers/simple-streamablehttp-stateless/
The streamable HTTP transport supports:
- Stateful and stateless operation modes
- Resumability with event stores
- JSON or SSE response formats
- Better scalability for multi-node deployments
Note: SSE transport is being superseded by Streamable HTTP transport.
By default, SSE servers are mounted at /sse
and Streamable HTTP servers are mounted at /mcp
. You can customize these paths using the methods described below.
You can mount the SSE server to an existing ASGI server using the sse_app
method. This allows you to integrate the SSE server with other ASGI applications.
from mcp.server.fastmcp import FastMCP app = FastMCP("Secure Server") # Tool requiring cryptographic request signatures @app.tool( etdi=True, etdi_require_request_signing=True, etdi_permissions=["banking:transfer"] ) def transfer_funds(amount: float, to_account: str) -> str: """High-security tool requiring signed requests""" return f"Transferred ${amount} to {to_account}" # Initialize request signing verification app.initialize_request_signing()
- Client generates RSA key pair automatically
- Signs tool invocation with private key
- Embeds signature in MCP request parameters (not transport headers)
- Server extracts signature from MCP request
- Verifies signature using client's public key
- Enforces in STRICT mode only
Request signing extends the MCP protocol itself using the extra="allow"
feature:
# Standard MCP request { "method": "tools/call", "params": { "name": "my_tool", "arguments": {"param": "value"} } } # ETDI signed request (backward compatible) { "method": "tools/call", "params": { "name": "my_tool", "arguments": {"param": "value"}, "etdi_signature": "base64-encoded-signature", "etdi_timestamp": "2024-01-01T12:00:00Z", "etdi_key_id": "client-key-id", "etdi_algorithm": "RS256" } }
This approach ensures full compatibility with all MCP transports (stdio, websocket, SSE) without requiring transport-layer modifications.
A simple server demonstrating resources, tools, and prompts:
from mcp.server.fastmcp import FastMCP mcp = FastMCP("Echo") @mcp.resource("echo://{message}") def echo_resource(message: str) -> str: """Echo a message as a resource""" return f"Resource echo: {message}" @mcp.tool() def echo_tool(message: str) -> str: """Echo a message as a tool""" return f"Tool echo: {message}" @mcp.prompt() def echo_prompt(message: str) -> str: """Create an echo prompt""" return f"Please process this message: {message}"
A more complex example showing database integration:
import sqlite3 from mcp.server.fastmcp import FastMCP mcp = FastMCP("SQLite Explorer") @mcp.resource("schema://main") def get_schema() -> str: """Provide the database schema as a resource""" conn = sqlite3.connect("database.db") schema = conn.execute("SELECT sql FROM sqlite_master WHERE type='table'").fetchall() return "\n".join(sql[0] for sql in schema if sql[0]) @mcp.tool() def query_data(sql: str) -> str: """Execute SQL queries safely""" conn = sqlite3.connect("database.db") try: result = conn.execute(sql).fetchall() return "\n".join(str(row) for row in result) except Exception as e: return f"Error: {str(e)}"
For more control, you can use the low-level server implementation directly. This gives you full access to the protocol and allows you to customize every aspect of your server, including lifecycle management through the lifespan API:
from contextlib import asynccontextmanager from collections.abc import AsyncIterator from fake_database import Database # Replace with your actual DB type from mcp.server import Server @asynccontextmanager async def server_lifespan(server: Server) -> AsyncIterator[dict]: """Manage server startup and shutdown lifecycle.""" # Initialize resources on startup db = await Database.connect() try: yield {"db": db} finally: # Clean up on shutdown await db.disconnect() # Pass lifespan to server server = Server("example-server", lifespan=server_lifespan) # Access lifespan context in handlers @server.call_tool() async def query_db(name: str, arguments: dict) -> list: ctx = server.request_context db = ctx.lifespan_context["db"] return await db.query(arguments["query"])
The lifespan API provides:
- A way to initialize resources when the server starts and clean them up when it stops
- Access to initialized resources through the request context in handlers
- Type-safe context passing between lifespan and request handlers
import mcp.server.stdio import mcp.types as types from mcp.server.lowlevel import NotificationOptions, Server from mcp.server.models import InitializationOptions # Create a server instance server = Server("example-server") @server.list_prompts() async def handle_list_prompts() -> list[types.Prompt]: return [ types.Prompt( name="example-prompt", description="An example prompt template", arguments=[ types.PromptArgument( name="arg1", description="Example argument", required=True ) ], ) ] @server.get_prompt() async def handle_get_prompt( name: str, arguments: dict[str, str] | None ) -> types.GetPromptResult: if name != "example-prompt": raise ValueError(f"Unknown prompt: {name}") return types.GetPromptResult( description="Example prompt", messages=[ types.PromptMessage( role="user", content=types.TextContent(type="text", text="Example prompt text"), ) ], ) async def run(): async with mcp.server.stdio.stdio_server() as (read_stream, write_stream): await server.run( read_stream, write_stream, InitializationOptions( server_name="example", server_version="0.1.0", capabilities=server.get_capabilities( notification_options=NotificationOptions(), experimental_capabilities={}, ), ), ) if __name__ == "__main__": import asyncio asyncio.run(run())
Caution: The mcp run
and mcp dev
tool doesn't support low-level server.
The SDK provides a high-level client interface for connecting to MCP servers using various transports:
from mcp import ClientSession, StdioServerParameters, types from mcp.client.stdio import stdio_client # Create server parameters for stdio connection server_params = StdioServerParameters( command="python", # Executable args=["example_server.py"], # Optional command line arguments env=None, # Optional environment variables ) # Optional: create a sampling callback async def handle_sampling_message( message: types.CreateMessageRequestParams, ) -> types.CreateMessageResult: return types.CreateMessageResult( role="assistant", content=types.TextContent( type="text", text="Hello, world! from model", ), model="gpt-3.5-turbo", stopReason="endTurn", ) async def run(): async with stdio_client(server_params) as (read, write): async with ClientSession( read, write, sampling_callback=handle_sampling_message ) as session: # Initialize the connection await session.initialize() # List available prompts prompts = await session.list_prompts() # Get a prompt prompt = await session.get_prompt( "example-prompt", arguments={"arg1": "value"} ) # List available resources resources = await session.list_resources() # List available tools tools = await session.list_tools() # Read a resource content, mime_type = await session.read_resource("file://some/path") # Call a tool result = await session.call_tool("tool-name", arguments={"arg1": "value"}) if __name__ == "__main__": import asyncio asyncio.run(run())
Clients can also connect using Streamable HTTP transport:
from mcp.client.streamable_http import streamablehttp_client from mcp import ClientSession async def main(): # Connect to a streamable HTTP server async with streamablehttp_client("example/mcp") as ( read_stream, write_stream, _, ): # Create a session using the client streams async with ClientSession(read_stream, write_stream) as session: # Initialize the connection await session.initialize() # Call a tool tool_result = await session.call_tool("echo", {"message": "hello"})
The SDK includes authorization support for connecting to protected MCP servers:
from mcp.client.auth import OAuthClientProvider, TokenStorage from mcp.client.session import ClientSession from mcp.client.streamable_http import streamablehttp_client from mcp.shared.auth import OAuthClientInformationFull, OAuthClientMetadata, OAuthToken class CustomTokenStorage(TokenStorage): """Simple in-memory token storage implementation.""" async def get_tokens(self) -> OAuthToken | None: pass async def set_tokens(self, tokens: OAuthToken) -> None: pass async def get_client_info(self) -> OAuthClientInformationFull | None: pass async def set_client_info(self, client_info: OAuthClientInformationFull) -> None: pass async def main(): # Set up OAuth authentication oauth_auth = OAuthClientProvider( server_url="https://api.example.com", client_metadata=OAuthClientMetadata( client_name="My Client", redirect_uris=["http://localhost:3000/callback"], grant_types=["authorization_code", "refresh_token"], response_types=["code"], ), storage=CustomTokenStorage(), redirect_handler=lambda url: print(f"Visit: {url}"), callback_handler=lambda: ("auth_code", None), ) # Use with streamable HTTP client async with streamablehttp_client( "https://api.example.com/mcp", auth=oauth_auth ) as (read, write, _): async with ClientSession(read, write) as session: await session.initialize() # Authenticated session ready
For a complete working example, see examples/clients/simple-auth-client/
.
The MCP protocol defines three core primitives that servers can implement:
Primitive | Control | Description | Example Use |
---|---|---|---|
Prompts | User-controlled | Interactive templates invoked by user choice | Slash commands, menu options |
Resources | Application-controlled | Contextual data managed by the client application | File contents, API responses |
Tools | Model-controlled | Functions exposed to the LLM to take actions | API calls, data updates |
MCP servers declare capabilities during initialization:
Capability | Feature Flag | Description |
---|---|---|
prompts | listChanged | Prompt template management |
resources | subscribe listChanged | Resource exposure and updates |
tools | listChanged | Tool discovery and execution |
logging | - | Server logging configuration |
completion | - | Argument completion suggestions |
- Model Context Protocol documentation
- Model Context Protocol specification
- Officially supported servers
- Fork the repository
- Create a feature branch
- Make your changes
- Add tests for new functionality
- Run the test suite
- Submit a pull request
MIT License - see LICENSE file for details.