Files
AutoGPT/autogpt_platform/backend/backend/blocks/mcp/client.py
Zamil Majdy f9f358c526 feat(mcp): Add MCP tool block with OAuth, tool discovery, and standard credential integration (#12011)
## Summary

<img width="1000" alt="image"
src="https://github.com/user-attachments/assets/18e8ef34-d222-453c-8b0a-1b25ef8cf806"
/>


<img width="250" alt="image"
src="https://github.com/user-attachments/assets/ba97556c-09c5-4f76-9f4e-49a2e8e57468"
/>

<img width="250" alt="image"
src="https://github.com/user-attachments/assets/68f7804a-fe74-442d-9849-39a229c052cf"
/>

<img width="250" alt="image"
src="https://github.com/user-attachments/assets/700690ba-f9fe-4726-8871-3bfbab586001"
/>

Full-stack MCP (Model Context Protocol) tool block integration that
allows users to connect to any MCP server, discover available tools,
authenticate via OAuth, and execute tools — all through the standard
AutoGPT credential system.

### Backend

- **MCPToolBlock** (`blocks/mcp/block.py`): New block using
`CredentialsMetaInput` pattern with optional credentials (`default={}`),
supporting both authenticated (OAuth) and public MCP servers. Includes
auto-lookup fallback for backward compatibility.
- **MCP Client** (`blocks/mcp/client.py`): HTTP transport with JSON-RPC
2.0, tool discovery, tool execution with robust error handling
(type-checked error fields, non-JSON response handling)
- **MCP OAuth Handler** (`blocks/mcp/oauth.py`): RFC 8414 discovery,
dynamic per-server OAuth with PKCE, token storage and refresh via
`raise_for_status=True`
- **MCP API Routes** (`api/features/mcp/routes.py`): `discover-tools`,
`oauth/login`, `oauth/callback` endpoints with credential cleanup,
defensive OAuth metadata validation
- **Credential system integration**:
- `CredentialsMetaInput` model_validator normalizes legacy
`"ProviderName.MCP"` format from Python 3.13's `str(StrEnum)` change
- `CredentialsFieldInfo.combine()` supports URL-based credential
discrimination (each MCP server gets its own credential entry)
- `aggregate_credentials_inputs` checks block schema defaults for
credential optionality
- Executor normalizes credential data for both Pydantic and JSON schema
validation paths
  - Chat credential matching handles MCP server URL filtering
- `provider_matches()` helper used consistently for Python 3.13 StrEnum
compatibility
- **Pre-run validation**: `_validate_graph_get_errors` now calls
`get_missing_input()` for custom block-level validation (MCP tool
arguments)
- **Security**: HTML tag stripping loop to prevent XSS bypass, SSRF
protection (removed trusted_origins)

### Frontend

- **MCPToolDialog** (`MCPToolDialog.tsx`): Full tool discovery UI —
enter server URL, authenticate if needed, browse tools, select tool and
configure
- **OAuth popup** (`oauth-popup.ts`): Shared utility supporting
cross-origin MCP OAuth flows with BroadcastChannel + localStorage
fallback
- **Credential integration**: MCP-specific OAuth flow in
`useCredentialsInput`, server URL filtering in `useCredentials`, MCP
callback page
- **CredentialsSelect**: Auto-selects first available credential instead
of defaulting to "None", credentials listed before "None" in dropdown
- **Node rendering**: Dynamic tool input schema rendering on MCP nodes,
proper handling in both legacy and new flow editors
- **Block title persistence**: `customized_name` set at block creation
for both MCP and Agent blocks — no fallback logic needed, titles survive
save/load reliably
- **Stable credential ordering**: Removed `sortByUnsetFirst` that caused
credential inputs to jump when selected

### Tests (~2060 lines)

- Unit tests: block, client, tool execution
- Integration tests: mock MCP server with auth
- OAuth flow tests
- API endpoint tests
- Credential combining/optionality tests
- E2e tests (skipped in CI, run manually)

## Key Design Decisions

1. **Optional credentials via `default={}`**: MCP servers can be public
(no auth) or private (OAuth). The `credentials` field has `default={}`
making it optional at the schema level, so public servers work without
prompting for credentials.

2. **URL-based credential discrimination**: Each MCP server URL gets its
own credential entry in the "Run agent" form (via
`discriminator="server_url"`), so agents using multiple MCP servers
prompt for each independently.

3. **Model-level normalization**: Python 3.13 changed `str(StrEnum)` to
return `"ClassName.MEMBER"`. Rather than scattering fixes across the
codebase, a Pydantic `model_validator(mode="before")` on
`CredentialsMetaInput` handles normalization centrally, and
`provider_matches()` handles lookups.

4. **Credential auto-select**: `CredentialsSelect` component defaults to
the first available credential and notifies the parent state, ensuring
credentials are pre-filled in the "Run agent" dialog without requiring
manual selection.

5. **customized_name for block titles**: Both MCP and Agent blocks set
`customized_name` in metadata at creation time. This eliminates
convoluted runtime fallback logic (`agent_name`, hostname extraction) —
the title is persisted once and read directly.

## Test plan

- [x] Unit/integration tests pass (68 MCP + 11 graph = 79 tests)
- [x] Manual: MCP block with public server (DeepWiki) — no credentials
needed, tools discovered and executable
- [x] Manual: MCP block with OAuth server (Linear, Sentry) — OAuth flow
prompts correctly
- [x] Manual: "Run agent" form shows correct credential requirements per
MCP server
- [x] Manual: Credential auto-selects when exactly one matches,
pre-selects first when multiple exist
- [x] Manual: Credential ordering stays stable when
selecting/deselecting
- [x] Manual: MCP block title persists after save and refresh
- [x] Manual: Agent block title persists after save and refresh (via
customized_name)
- [ ] Manual: Shared agent with MCP block prompts new user for
credentials

---------

Co-authored-by: Otto <otto@agpt.co>
Co-authored-by: Ubbe <hi@ubbe.dev>
2026-02-13 16:17:03 +00:00

324 lines
11 KiB
Python

"""
MCP (Model Context Protocol) HTTP client.
Implements the MCP Streamable HTTP transport for listing tools and calling tools
on remote MCP servers. Uses JSON-RPC 2.0 over HTTP POST.
Handles both JSON and SSE (text/event-stream) response formats per the MCP spec.
Reference: https://modelcontextprotocol.io/specification/2025-03-26/basic/transports
"""
import json
import logging
from dataclasses import dataclass, field
from typing import Any
from backend.util.request import Requests
logger = logging.getLogger(__name__)
@dataclass
class MCPTool:
"""Represents an MCP tool discovered from a server."""
name: str
description: str
input_schema: dict[str, Any]
@dataclass
class MCPCallResult:
"""Result from calling an MCP tool."""
content: list[dict[str, Any]] = field(default_factory=list)
is_error: bool = False
class MCPClientError(Exception):
"""Raised when an MCP protocol error occurs."""
pass
class MCPClient:
"""
Async HTTP client for the MCP Streamable HTTP transport.
Communicates with MCP servers using JSON-RPC 2.0 over HTTP POST.
Supports optional Bearer token authentication.
"""
def __init__(
self,
server_url: str,
auth_token: str | None = None,
):
self.server_url = server_url.rstrip("/")
self.auth_token = auth_token
self._request_id = 0
self._session_id: str | None = None
def _next_id(self) -> int:
self._request_id += 1
return self._request_id
def _build_headers(self) -> dict[str, str]:
headers = {
"Content-Type": "application/json",
"Accept": "application/json, text/event-stream",
}
if self.auth_token:
headers["Authorization"] = f"Bearer {self.auth_token}"
if self._session_id:
headers["Mcp-Session-Id"] = self._session_id
return headers
def _build_jsonrpc_request(
self, method: str, params: dict[str, Any] | None = None
) -> dict[str, Any]:
req: dict[str, Any] = {
"jsonrpc": "2.0",
"method": method,
"id": self._next_id(),
}
if params is not None:
req["params"] = params
return req
@staticmethod
def _parse_sse_response(text: str) -> dict[str, Any]:
"""Parse an SSE (text/event-stream) response body into JSON-RPC data.
MCP servers may return responses as SSE with format:
event: message
data: {"jsonrpc":"2.0","result":{...},"id":1}
We extract the last `data:` line that contains a JSON-RPC response
(i.e. has an "id" field), which is the reply to our request.
"""
last_data: dict[str, Any] | None = None
for line in text.splitlines():
stripped = line.strip()
if stripped.startswith("data:"):
payload = stripped[len("data:") :].strip()
if not payload:
continue
try:
parsed = json.loads(payload)
# Only keep JSON-RPC responses (have "id"), skip notifications
if isinstance(parsed, dict) and "id" in parsed:
last_data = parsed
except (json.JSONDecodeError, ValueError):
continue
if last_data is None:
raise MCPClientError("No JSON-RPC response found in SSE stream")
return last_data
async def _send_request(
self, method: str, params: dict[str, Any] | None = None
) -> Any:
"""Send a JSON-RPC request to the MCP server and return the result.
Handles both ``application/json`` and ``text/event-stream`` responses
as required by the MCP Streamable HTTP transport specification.
"""
payload = self._build_jsonrpc_request(method, params)
headers = self._build_headers()
requests = Requests(
raise_for_status=True,
extra_headers=headers,
)
response = await requests.post(self.server_url, json=payload)
# Capture session ID from response (MCP Streamable HTTP transport)
session_id = response.headers.get("Mcp-Session-Id")
if session_id:
self._session_id = session_id
content_type = response.headers.get("content-type", "")
if "text/event-stream" in content_type:
body = self._parse_sse_response(response.text())
else:
try:
body = response.json()
except Exception as e:
raise MCPClientError(
f"MCP server returned non-JSON response: {e}"
) from e
if not isinstance(body, dict):
raise MCPClientError(
f"MCP server returned unexpected JSON type: {type(body).__name__}"
)
# Handle JSON-RPC error
if "error" in body:
error = body["error"]
if isinstance(error, dict):
raise MCPClientError(
f"MCP server error [{error.get('code', '?')}]: "
f"{error.get('message', 'Unknown error')}"
)
raise MCPClientError(f"MCP server error: {error}")
return body.get("result")
async def _send_notification(self, method: str) -> None:
"""Send a JSON-RPC notification (no id, no response expected)."""
headers = self._build_headers()
notification = {"jsonrpc": "2.0", "method": method}
requests = Requests(
raise_for_status=False,
extra_headers=headers,
)
await requests.post(self.server_url, json=notification)
async def discover_auth(self) -> dict[str, Any] | None:
"""Probe the MCP server's OAuth metadata (RFC 9728 / MCP spec).
Returns ``None`` if the server doesn't require auth, otherwise returns
a dict with:
- ``authorization_servers``: list of authorization server URLs
- ``resource``: the resource indicator URL (usually the MCP endpoint)
- ``scopes_supported``: optional list of supported scopes
The caller can then fetch the authorization server metadata to get
``authorization_endpoint``, ``token_endpoint``, etc.
"""
from urllib.parse import urlparse
parsed = urlparse(self.server_url)
base = f"{parsed.scheme}://{parsed.netloc}"
# Build candidates for protected-resource metadata (per RFC 9728)
path = parsed.path.rstrip("/")
candidates = []
if path and path != "/":
candidates.append(f"{base}/.well-known/oauth-protected-resource{path}")
candidates.append(f"{base}/.well-known/oauth-protected-resource")
requests = Requests(
raise_for_status=False,
)
for url in candidates:
try:
resp = await requests.get(url)
if resp.status == 200:
data = resp.json()
if isinstance(data, dict) and "authorization_servers" in data:
return data
except Exception:
continue
return None
async def discover_auth_server_metadata(
self, auth_server_url: str
) -> dict[str, Any] | None:
"""Fetch the OAuth Authorization Server Metadata (RFC 8414).
Given an authorization server URL, returns a dict with:
- ``authorization_endpoint``
- ``token_endpoint``
- ``registration_endpoint`` (for dynamic client registration)
- ``scopes_supported``
- ``code_challenge_methods_supported``
- etc.
"""
from urllib.parse import urlparse
parsed = urlparse(auth_server_url)
base = f"{parsed.scheme}://{parsed.netloc}"
path = parsed.path.rstrip("/")
# Try standard metadata endpoints (RFC 8414 and OpenID Connect)
candidates = []
if path and path != "/":
candidates.append(f"{base}/.well-known/oauth-authorization-server{path}")
candidates.append(f"{base}/.well-known/oauth-authorization-server")
candidates.append(f"{base}/.well-known/openid-configuration")
requests = Requests(
raise_for_status=False,
)
for url in candidates:
try:
resp = await requests.get(url)
if resp.status == 200:
data = resp.json()
if isinstance(data, dict) and "authorization_endpoint" in data:
return data
except Exception:
continue
return None
async def initialize(self) -> dict[str, Any]:
"""
Send the MCP initialize request.
This is required by the MCP protocol before any other requests.
Returns the server's capabilities.
"""
result = await self._send_request(
"initialize",
{
"protocolVersion": "2025-03-26",
"capabilities": {},
"clientInfo": {"name": "AutoGPT-Platform", "version": "1.0.0"},
},
)
# Send initialized notification (no response expected)
await self._send_notification("notifications/initialized")
return result or {}
async def list_tools(self) -> list[MCPTool]:
"""
Discover available tools from the MCP server.
Returns a list of MCPTool objects with name, description, and input schema.
"""
result = await self._send_request("tools/list")
if not result or "tools" not in result:
return []
tools = []
for tool_data in result["tools"]:
tools.append(
MCPTool(
name=tool_data.get("name", ""),
description=tool_data.get("description", ""),
input_schema=tool_data.get("inputSchema", {}),
)
)
return tools
async def call_tool(
self, tool_name: str, arguments: dict[str, Any]
) -> MCPCallResult:
"""
Call a tool on the MCP server.
Args:
tool_name: The name of the tool to call.
arguments: The arguments to pass to the tool.
Returns:
MCPCallResult with the tool's response content.
"""
result = await self._send_request(
"tools/call",
{"name": tool_name, "arguments": arguments},
)
if not result:
return MCPCallResult(is_error=True)
return MCPCallResult(
content=result.get("content", []),
is_error=result.get("isError", False),
)