Files
cp1-agents/mcp-servers/aws-discovery/server.py
Jonathan Boniface c14a94f33b
Some checks failed
check agents / check (push) Failing after 12s
Add hyperscaler discovery MCP servers and Well-Architected guidance
azure-discovery, aws-discovery and gcp-discovery collect read-only metadata
on servers, workloads, Kubernetes clusters, networks, data stores, policies
and provider recommendations. Credentials arrive per call from the vault.
Replies are paged, scrubbed of secrets and anonymised, and throttled calls
back off. cloud-knowledge searches each provider's Well-Architected guidance
from fixed lists only, so no customer detail can leave the platform.

Adds shared skills, the render and test tooling, a public Python Dockerfile
until the hardened base image is reachable, the customer access guide, the
implementation plan and the Vault Service platform issue.

Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
2026-10-30 20:11:57 +00:00

1459 lines
69 KiB
Python

"""AWS Discovery
=============
Read-only discovery of a customer's AWS estate - servers, workloads, Kubernetes clusters, networks, data stores, policies and Trusted Advisor findings.
Generated by Agent Builder Platform - MCP Server Builder
Version: 0.1.1
Template: mcp-server-template 1.1.0
Features:
- JSON-RPC 2.0 over HTTP (MCP standard)
- Bearer token authentication
- Health check endpoint
- Structured logging (JSON, one object per line)
Environment Variables:
- MCP_REGISTRATION_API_KEY: Authentication key for Gateway registration (required)
- MCP_SERVER_PORT: Server port (default: 8112)
- MCP_SERVER_NAME: Display name for the server
- LOG_LEVEL: Logging level (default: INFO)
- LOG_FORMAT: json (default) or console
"""
import os
import re
import sys
import json
import inspect
import time
import traceback
import asyncio
import logging
import secrets
from typing import Dict, Any, List, Optional, Union # noqa: F401
from collections import OrderedDict
from contextlib import asynccontextmanager
from datetime import datetime, timezone
import structlog
from fastapi import FastAPI, HTTPException, Depends, Request
from fastapi.security import HTTPBearer, HTTPAuthorizationCredentials
from pydantic import BaseModel, Field, StrictInt, StrictStr
import mcp_connection
# Additional imports from tool implementations
import base64
import hashlib
import boto3
from botocore.config import Config as BotoConfig
TEMPLATE_VERSION = '1.1.0'
# What this server needs to reach its upstream, advertised to the gateway on initialize (None: nothing)
CONNECTION_SCHEMA = {'type': 'object',
'title': 'AWS Discovery',
'description': "Read-only discovery of a customer's AWS estate - servers, workloads, "
'Kubernetes clusters, networks, data stores, policies and Trusted '
'Advisor findings.',
'properties': {'access_key_id': {'type': 'string',
'title': 'Access key ID',
'description': 'The broker identity that assumes the '
'read-only role.'},
'secret_access_key': {'type': 'string',
'title': 'Secret access key',
'description': "The broker identity's secret "
'key.',
'writeOnly': True,
'format': 'password'},
'role_arn': {'type': 'string',
'title': 'Role ARN',
'description': "The customer's read-only role."},
'external_id': {'type': 'string',
'title': 'External ID',
'description': "The external id on the role's trust "
'policy.',
'writeOnly': True,
'format': 'password'},
'regions': {'type': 'string',
'title': 'Regions',
'description': 'Comma-separated regions that may be read. '
'The first is the default.'},
'account_ids': {'type': 'string',
'title': 'Account IDs',
'description': 'Comma-separated member accounts that '
'may be read.'},
'member_role_name': {'type': 'string',
'title': 'Member role name',
'description': "The read-only role's name in "
'member accounts.'},
'aggregator_name': {'type': 'string',
'title': 'Config aggregator',
'description': 'An AWS Config aggregator for '
'organisation-wide queries.'}},
'required': ['access_key_id',
'secret_access_key',
'role_arn',
'external_id',
'regions'],
'additionalProperties': False}
VERIFY_TOOL = 'verify_connection'
# ============================================
# Configuration
# ============================================
def _load_env_file() -> None:
"""Read the `.env` next to this file, so `cp .env.example .env` and `python server.py` work.
Variables that are already set win, so a container's or a pipeline's settings are never overridden.
The image does not contain a `.env`, so this does nothing there.
"""
path = globals().get("__file__")
if not path:
return
try:
from dotenv import load_dotenv
except ImportError:
return
load_dotenv(os.path.join(os.path.dirname(os.path.abspath(path)), ".env"))
_load_env_file()
class Config:
"""Server configuration from environment variables"""
# Server Identity
SERVER_ID = os.getenv("MCP_SERVER_ID", 'aws-discovery')
SERVER_NAME = os.getenv("MCP_SERVER_NAME", 'AWS Discovery')
SERVER_VERSION = os.getenv("MCP_SERVER_VERSION", '0.1.1')
# MCP_SERVER_PORT, or SERVER_PORT as the ConfigMaps in the tools repo name it
SERVER_PORT = int(os.getenv("MCP_SERVER_PORT") or os.getenv("SERVER_PORT") or "8112")
# Authentication (REQUIRED for MCP Gateway registration)
# Generate secure key: openssl rand -hex 32
MCP_REGISTRATION_API_KEY = os.getenv("MCP_REGISTRATION_API_KEY", "").strip()
# The placeholder from .env.example is not a key: a server started with it unchanged has no key.
if MCP_REGISTRATION_API_KEY.startswith("<"):
MCP_REGISTRATION_API_KEY = ""
# Feature Flags
REQUIRE_AUTH = os.getenv("REQUIRE_AUTH", "true").lower() == "true"
# Applies when CONNECTION_SCHEMA is set: refuse a call that arrives without the tenant's connection
# instead of falling back to this server's own environment. Set to false only for local runs.
REQUIRE_VAULT_CONNECTION = os.getenv("REQUIRE_VAULT_CONNECTION", "true").lower() == "true"
# Logging: LOG_FORMAT is "json" (one object per line, for log stores) or "console" (for reading by eye)
LOG_LEVEL = os.getenv("LOG_LEVEL", "INFO").upper()
LOG_FORMAT = os.getenv("LOG_FORMAT", "json").lower()
config = Config()
# ============================================
# Logging Configuration
# ============================================
def sanitize_log(value: Any, max_length: int = 200) -> str:
"""Make a caller-supplied value safe to log: one line, no control characters, bounded."""
if not isinstance(value, str):
value = str(value)
# Written with .replace() because that is the pattern CodeQL recognises as a
# barrier for py/log-injection (CWE-117).
value = value.replace("\n", "").replace("\r", "")
return re.sub(r"[\x00-\x08\x0b\x0c\x0e-\x1f\x7f-\x9f]", "", value)[:max_length]
def _add_server_identity(_logger, _method, event_dict):
event_dict.setdefault("server_id", config.SERVER_ID)
event_dict.setdefault("server_name", config.SERVER_NAME)
event_dict.setdefault("server_version", config.SERVER_VERSION)
return event_dict
_LEADING_KEYS = (
"timestamp", "level", "event", "logger", "server_id", "server_name", "server_version",
"request_id", "method", "tool",
)
def _scrub_secrets(_logger, _method, event_dict):
"""The last step before a line is written: no credential reaches the log, whoever logged it."""
return mcp_connection.scrub(event_dict)
def _lead_with_common_keys(_logger, _method, event_dict):
"""Every line starts with the same fields in the same order; the event's own follow."""
lead = {key: event_dict.pop(key) for key in _LEADING_KEYS if key in event_dict}
lead.update(event_dict)
return lead
def configure_logging() -> None:
"""Send every log line, ours and the libraries', to stdout in one format."""
level = getattr(logging, config.LOG_LEVEL, logging.INFO)
shared = [
structlog.contextvars.merge_contextvars,
structlog.stdlib.add_logger_name,
structlog.stdlib.add_log_level,
structlog.processors.TimeStamper(fmt="iso", utc=True),
_add_server_identity,
]
renderer = (
structlog.dev.ConsoleRenderer(colors=False)
if config.LOG_FORMAT == "console"
else structlog.processors.JSONRenderer()
)
structlog.configure(
processors=shared + [
structlog.stdlib.PositionalArgumentsFormatter(),
structlog.processors.StackInfoRenderer(),
structlog.processors.format_exc_info,
structlog.stdlib.ProcessorFormatter.wrap_for_formatter,
],
logger_factory=structlog.stdlib.LoggerFactory(),
wrapper_class=structlog.stdlib.BoundLogger,
cache_logger_on_first_use=True,
)
handler = logging.StreamHandler(sys.stdout)
handler.mcp_server_handler = True
handler.setFormatter(structlog.stdlib.ProcessorFormatter(
foreign_pre_chain=shared,
processors=[
structlog.stdlib.ProcessorFormatter.remove_processors_meta,
structlog.processors.format_exc_info,
_lead_with_common_keys,
_scrub_secrets,
renderer,
],
))
root = logging.getLogger()
root.handlers = [h for h in root.handlers if not getattr(h, "mcp_server_handler", False)] + [handler]
root.setLevel(level)
# httpx logs every request URL at INFO, and an upstream URL often carries a key: keep those lines for DEBUG
for name in ("httpx", "httpcore"):
logging.getLogger(name).setLevel(level if level <= logging.DEBUG else logging.WARNING)
# uvicorn installs its own handlers; route them through ours so every line has the same shape
for name in ("uvicorn", "uvicorn.error", "uvicorn.access"):
uvicorn_logger = logging.getLogger(name)
uvicorn_logger.handlers.clear()
uvicorn_logger.propagate = True
configure_logging()
logger = structlog.get_logger(__name__)
# ============================================
# Per-request connection
# ============================================
CONNECTION_METHODS = ("tools/call", "resources/read", "prompts/get")
def get_connection() -> Dict[str, Any]:
"""The connection fields the gateway sent with this request; empty when it sent none."""
return dict(mcp_connection.active() or {})
_client_cache: "OrderedDict[str, Any]" = OrderedDict()
_CLIENT_CACHE_MAX = 32
def get_upstream_client(kind, build):
"""The upstream client of type `kind` for this request's connection.
`build(connection)` runs once per distinct (kind, connection), so two tools that talk to different
upstreams with the same tenant connection each get their own client.
"""
if not isinstance(kind, str) or not kind:
raise TypeError("get_upstream_client needs a client kind, for example get_upstream_client('jira', build)")
key = f"{kind}:{mcp_connection.fingerprint(mcp_connection.active())}"
client = _client_cache.get(key)
if client is None:
client = build(get_connection())
_client_cache[key] = client
while len(_client_cache) > _CLIENT_CACHE_MAX:
_close_client(_client_cache.popitem(last=False)[1])
else:
_client_cache.move_to_end(key)
return client
_closing_tasks: set = set()
def _close_client(client) -> None:
"""Release the connections of a client that left the cache (an httpx client holds a pool of sockets).
Closes with `aclose()`, `close()` or `dispose()` (an httpx client, most clients, a SQLAlchemy engine); a client
with none of them is left to the garbage collector. A failure to close is logged and never reaches the
request that happened to evict it.
"""
close = getattr(client, "aclose", None) or getattr(client, "close", None) or getattr(client, "dispose", None)
if close is None:
return
try:
result = close()
if inspect.isawaitable(result):
try:
task = asyncio.get_running_loop().create_task(result)
except RuntimeError: # no loop to run it on
result.close()
return
_closing_tasks.add(task) # a task nobody refers to can be collected before it finishes
task.add_done_callback(_closing_tasks.discard)
except Exception as error:
logger.warning("client_close_failed", error_type=type(error).__name__, error=sanitize_log(error))
async def _close_all_clients() -> None:
"""On shutdown: close every cached client and wait for the closes already under way."""
while _client_cache:
_close_client(_client_cache.popitem(last=False)[1])
if _closing_tasks:
await asyncio.gather(*list(_closing_tasks), return_exceptions=True)
def check_connection(profile: Optional[Dict[str, Any]]) -> None:
"""Refuse a call with no connection (unless the environment may be used) or with an incomplete one."""
if not profile:
if config.REQUIRE_VAULT_CONNECTION:
raise ValueError(
"No connection was sent with this request: no credential is stored for this tenant. "
"Store one under Credentials, or set REQUIRE_VAULT_CONNECTION=false to use this "
"server's own environment."
)
return
missing = mcp_connection.missing_fields(profile, CONNECTION_SCHEMA)
if missing:
raise ValueError("The connection sent with this request is missing: " + ", ".join(missing))
def render_prompt(template: str, values: Dict[str, Any]) -> str:
"""Fill {{name}} markers in a prompt template; a marker with no value is left as written."""
return re.sub(
r"\{\{\s*([A-Za-z_][A-Za-z0-9_]*)\s*\}\}",
lambda m: str(values[m.group(1)]) if m.group(1) in values else m.group(0),
template,
)
# ============================================
# Authentication
# ============================================
security = HTTPBearer(auto_error=False)
async def verify_mcp_registration_key(
credentials: Optional[HTTPAuthorizationCredentials] = Depends(security)
) -> bool:
"""
Verify MCP registration API key for /messages endpoint.
This is the industry-standard Bearer token authentication pattern.
"""
# Skip auth if disabled (for testing only!)
if not config.REQUIRE_AUTH:
logger.warning("auth_disabled", detail="testing mode only")
return True
if not credentials:
logger.warning("auth_missing_credentials")
raise HTTPException(
status_code=401,
detail="Missing authentication credentials. Include 'Authorization: Bearer <token>' header.",
headers={"WWW-Authenticate": "Bearer"}
)
provided_key = credentials.credentials
expected_key = config.MCP_REGISTRATION_API_KEY
if not expected_key:
logger.error("auth_key_not_configured")
raise HTTPException(
status_code=500,
detail="Server authentication not configured"
)
if not secrets.compare_digest(provided_key.encode("utf-8"), expected_key.encode("utf-8")):
logger.warning("auth_invalid_key")
raise HTTPException(
status_code=401,
detail="Invalid authentication credentials",
headers={"WWW-Authenticate": "Bearer"}
)
return True
# ============================================
# MCP Protocol Models
# ============================================
class MCPRequest(BaseModel):
"""JSON-RPC 2.0 request (MCP standard)"""
jsonrpc: str = "2.0"
id: Union[StrictStr, StrictInt] # JSON-RPC allows a string or a number; the reply echoes it unchanged
method: str
params: Dict[str, Any] = Field(default_factory=dict)
class MCPResponse(BaseModel):
"""JSON-RPC 2.0 response (MCP standard)"""
jsonrpc: str = "2.0"
id: Union[StrictStr, StrictInt]
result: Optional[Dict[str, Any]] = None
error: Optional[Dict[str, Any]] = None
# ============================================
# Tool Definitions
# ============================================
TOOLS_REGISTRY = [
{
"name": 'aws_check_connection',
"description": 'Test the stored connection: assume the role and read its identity. Returns no customer data.',
"inputSchema": {'type': 'object', 'properties': {}, 'required': []}
},
{
"name": 'aws_list_accounts',
"description": "List the organisation's accounts or organizational units. Needs the management or a delegated administrator account.",
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string', 'description': 'One of: accounts, roots.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_resources',
"description": 'List resources of one type that AWS Config has recorded.',
"inputSchema": {'type': 'object',
'properties': {'resource_type': {'type': 'string',
'description': 'A Config type, such as '
'AWS::EC2::Instance.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': ['resource_type']}
},
{
"name": 'aws_query_resources',
"description": 'Run a read-only AWS Config advanced query. Uses the organisation aggregator when the connection names one.',
"inputSchema": {'type': 'object',
'properties': {'query': {'type': 'string',
'description': 'A Config SELECT expression.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': ['query']}
},
{
"name": 'aws_get_resource_metadata',
"description": 'Read-only provider metadata for one resource, from AWS Config.',
"inputSchema": {'type': 'object',
'properties': {'resource_type': {'type': 'string', 'description': 'A Config type.'},
'resource_id': {'type': 'string', 'description': 'The resource id.'},
'scope_entry_id': {'type': 'string',
'description': 'The scope entry this answers, '
'echoed in the reply.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'}},
'required': ['resource_type', 'resource_id']}
},
{
"name": 'aws_list_servers',
"description": 'List EC2 instances, Auto Scaling groups or EBS volumes.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: instances, auto_scaling_groups, '
'volumes.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_workloads',
"description": 'List Lambda functions, ECS clusters, App Runner services or Elastic Beanstalk environments. Environment variables are removed.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: lambda_functions, ecs_clusters, '
'app_runner_services, '
'elastic_beanstalk_environments.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_kubernetes_clusters',
"description": 'List EKS clusters with version, node groups and networking. Stops at the cloud control plane.',
"inputSchema": {'type': 'object',
'properties': {'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_get_network_topology',
"description": 'List VPCs, subnets, route tables, security groups, NAT or internet gateways, transit gateways, peerings, endpoints or load balancers.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: vpcs, subnets, route_tables, '
'security_groups, nat_gateways, '
'internet_gateways, transit_gateways, '
'vpc_peering, vpc_endpoints, load_balancers.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_data_stores',
"description": 'List RDS instances, DynamoDB tables or S3 buckets. Metadata only, never rows or objects.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: rds_instances, dynamodb_tables, '
's3_buckets.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_organization_policies',
"description": 'List Organizations policies with their documents, or Config rule compliance.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: service_control_policies, '
'resource_control_policies, tag_policies, '
'backup_policies, config_rule_compliance.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_resource_policies',
"description": 'List resource-based policies: S3 buckets, KMS keys, Lambda functions, SQS queues, SNS topics, or IAM role trust policies. People are anonymised.',
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: s3_buckets, kms_keys, '
'lambda_functions, sqs_queues, sns_topics, '
'iam_role_trust.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'aws_list_recommendations',
"description": "List Trusted Advisor recommendations, each with a common Well-Architected pillar, or the customer's existing Well-Architected Tool reviews.",
"inputSchema": {'type': 'object',
'properties': {'kind': {'type': 'string',
'description': 'One of: trusted_advisor, '
'well_architected_reviews.'},
'account_id': {'type': 'string',
'description': "An account from the connection's list. "
"Default: the connection's own."},
'region': {'type': 'string',
'description': "A region from the connection's list. "
'Default: the first.'},
'page_size': {'type': 'integer',
'description': 'Items per page, up to 200. Default 50.'},
'page_token': {'type': 'string',
'description': 'next_page_token from the previous '
'reply.'}},
'required': []}
},
{
"name": 'verify_connection',
"description": (
"Check that the connection sent with this request works against the upstream, and say who it "
"authenticated as. Makes no change."
),
"inputSchema": {'type': 'object', 'properties': {}, 'required': []}
}
]
# ============================================
# Resource Definitions
# ============================================
RESOURCES_REGISTRY = []
# ============================================
# Prompt Definitions
# ============================================
PROMPT_TEMPLATES = {}
PROMPTS_REGISTRY = [
{
"name": name,
"description": prompt["description"],
"arguments": [
{"name": a["name"], "description": a["description"], "required": a["required"]}
for a in prompt["arguments"]
],
}
for name, prompt in PROMPT_TEMPLATES.items()
]
# ============================================
# Tool Implementations
# ============================================
DEFAULT_PAGE_SIZE = 50
MAX_PAGE_SIZE = 200
MAX_RESPONSE_BYTES = 200000
RETRY_ATTEMPTS = 5
RETRY_MAX_WAIT_SECONDS = 20
RETRY_BUDGET_SECONDS = 60
SECRET_KEYS = (
"userdata", "customdata", "customscript", "protectedsettings", "connectionstring", "connectionstrings",
"appsettings", "env", "envvars", "environmentvariables", "variables", "password", "adminpassword", "secret",
"secretvalue", "sastoken", "accesskey", "primarykey", "secondarykey", "privatekey", "kubeconfig", "credentials",
)
PERSON_KEYS = (
"createdby", "lastmodifiedby", "updatedby", "creator", "owner", "owneremail", "contact", "contactemail", "email",
"objectid", "principalid", "principalname", "userprincipalname",
)
PRINCIPAL_KINDS = {"user": "[user]", "group": "[group]", "domain": "[domain]", "principal": "[user]",
"principalSet": "[group]", "deleted": "[user]"}
NON_PERSON_EMAIL_DOMAINS = ("gserviceaccount.com",)
def _clean_text(text: str) -> str:
"""Anonymise people inside a string: emails, AWS IAM users, and session names, which are often emails."""
def email(match):
return match.group(0) if match.group(0).lower().endswith(NON_PERSON_EMAIL_DOMAINS) else "[email]"
text = re.sub(r"arn:aws[a-z-]*:iam::\d{12}:user/[^\s\"',]+", "[user]", text)
text = re.sub(r"(arn:aws[a-z-]*:sts::\d{12}:assumed-role/[^/\s\"',]+)/[^\s\"',]+", r"\1/[session]", text)
# Bounded, and skipped without an @: an unbounded pattern takes minutes on one long string.
if "@" not in text:
return text
return re.sub(r"[A-Za-z0-9._%+-]{1,64}@[A-Za-z0-9.-]{1,255}\.[A-Za-z]{2,24}", email, text)
def _anonymise_member(value: str) -> str:
"""One IAM member, such as user:a@b.example. Service accounts stay: Discovery shows what can reach what."""
kind = value.split(":", 1)[0]
if ":" in value and kind == "serviceAccount":
return value
if ":" in value and kind in PRINCIPAL_KINDS:
return PRINCIPAL_KINDS[kind]
return _clean_text(value)
def _scrub(value, key: str = ""):
"""Drop secrets and anonymise people, however deep. Every tool's reply passes through here."""
lowered = key.lower().replace("_", "").replace("-", "")
if lowered in SECRET_KEYS:
return "[redacted]"
if lowered in PERSON_KEYS:
return "[user]"
if isinstance(value, dict):
tag = str(value.get("Key", value.get("key", ""))).lower().replace("_", "").replace("-", "")
cleaned = {k: _scrub(v, str(k)) for k, v in value.items()}
for name in ("Value", "value"):
if name in cleaned and tag in PERSON_KEYS:
cleaned[name] = "[user]"
elif name in cleaned and tag in SECRET_KEYS:
cleaned[name] = "[redacted]"
return cleaned
if isinstance(value, (list, tuple)):
return [_scrub(v, "members" if lowered == "members" else "") for v in value]
if isinstance(value, str):
return _anonymise_member(value) if lowered == "members" else _clean_text(value)
if value is None or isinstance(value, (bool, int, float)):
return value
return str(value)
def _retry_delay(response, attempt: int) -> float:
"""How long the provider asked us to wait: Retry-After, or Resource Graph's hh:mm:ss quota reset."""
reset = response.headers.get("x-ms-user-quota-resets-after", "")
if reset.count(":") == 2:
hours, minutes, seconds = reset.split(":")
return int(hours) * 3600 + int(minutes) * 60 + float(seconds)
try:
return float(response.headers.get("retry-after", ""))
except ValueError:
return float(2 ** attempt)
async def _with_backoff(send):
"""Send, waiting when the provider throttles. The waits stay well inside the 120-second limit on an MCP call."""
waited = 0.0
response = await send()
for attempt in range(RETRY_ATTEMPTS):
if response.status_code not in (429, 503):
return response
delay = min(_retry_delay(response, attempt), RETRY_MAX_WAIT_SECONDS)
if waited + delay > RETRY_BUDGET_SECONDS:
return response
logger.info("upstream_throttled", status=response.status_code, wait_seconds=delay)
await asyncio.sleep(delay)
waited += delay
response = await send()
return response
def _page_size(arguments: Dict[str, Any]) -> int:
try:
size = int(arguments.get("page_size") or DEFAULT_PAGE_SIZE)
except (TypeError, ValueError):
return DEFAULT_PAGE_SIZE
return max(1, min(size, MAX_PAGE_SIZE))
def _read_page(arguments: Dict[str, Any]):
"""(upstream token, items of that upstream page already returned)."""
token = arguments.get("page_token")
if not token:
return None, 0
state = json.loads(base64.urlsafe_b64decode(str(token).encode()).decode())
return state["u"], int(state["s"])
def _write_page(upstream, skip: int) -> str:
return base64.urlsafe_b64encode(json.dumps({"u": upstream, "s": skip}).encode()).decode()
def _evidence_ref(tool: str, arguments: Dict[str, Any], payload) -> str:
asked = {k: v for k, v in arguments.items() if k != "page_token"}
blob = json.dumps([tool, asked, payload], sort_keys=True, default=str)
return "sha256:" + hashlib.sha256(blob.encode()).hexdigest()
def _collected_at() -> str:
return time.strftime("%Y-%m-%dT%H:%M:%SZ", time.gmtime())
def _refuse(code: str, message: str) -> Dict[str, Any]:
return {"success": False, "error": {"code": code, "message": message}}
def _upstream_failed(provider: str, error: Exception) -> Dict[str, Any]:
"""Name the failure and never its text: an upstream error can quote a request or an identity."""
status = getattr(getattr(error, "response", None), "status_code", None)
code = getattr(error, "response", None)
if isinstance(code, dict):
status = code.get("Error", {}).get("Code")
logger.warning("upstream_call_failed", provider=provider, error_type=type(error).__name__, status=status)
detail = f" ({status})" if status else ""
return _refuse("upstream_failed", f"{provider} refused or failed the call: {type(error).__name__}{detail}")
def _page_result(provider: str, tool: str, arguments: Dict[str, Any], items: List[Any], used, upstream_next,
skip: int = 0, extra: Optional[Dict[str, Any]] = None) -> Dict[str, Any]:
"""One page of scrubbed items, cut at MAX_RESPONSE_BYTES; the token resumes exactly where the cut fell."""
remaining = [_scrub(item) for item in items[skip:]]
kept, size = [], 0
for item in remaining:
size += len(json.dumps(item, default=str))
if kept and size > MAX_RESPONSE_BYTES:
break
kept.append(item)
if len(kept) < len(remaining):
next_token = _write_page(used, skip + len(kept))
else:
next_token = _write_page(upstream_next, 0) if upstream_next else None
data = {"provider": provider, "items": kept, "count": len(kept), "next_page_token": next_token,
"collected_at": _collected_at(), "evidence_ref": _evidence_ref(tool, arguments, kept)}
data.update(extra or {})
return {"success": True, "data": data}
def _one_result(provider: str, tool: str, arguments: Dict[str, Any], metadata: Any) -> Dict[str, Any]:
clean = _scrub(metadata)
return {"success": True, "data": {"provider": provider, "metadata": clean, "collected_at": _collected_at(),
"evidence_ref": _evidence_ref(tool, arguments, clean)}}
AWS_SESSIONS = {}
AWS_SERVERS = {
"instances": ("ec2", "describe_instances", "Reservations", "NextToken", "NextToken", {}),
"auto_scaling_groups": ("autoscaling", "describe_auto_scaling_groups", "AutoScalingGroups", "NextToken", "NextToken", {}),
"volumes": ("ec2", "describe_volumes", "Volumes", "NextToken", "NextToken", {}),
}
AWS_WORKLOADS = {
"lambda_functions": ("lambda", "list_functions", "Functions", "Marker", "NextMarker", {}),
"ecs_clusters": ("ecs", "list_clusters", "clusterArns", "nextToken", "nextToken", {}),
"app_runner_services": ("apprunner", "list_services", "ServiceSummaryList", "NextToken", "NextToken", {}),
"elastic_beanstalk_environments": ("elasticbeanstalk", "describe_environments", "Environments", "NextToken", "NextToken", {}),
}
AWS_NETWORK = {
"vpcs": ("ec2", "describe_vpcs", "Vpcs", "NextToken", "NextToken", {}),
"subnets": ("ec2", "describe_subnets", "Subnets", "NextToken", "NextToken", {}),
"route_tables": ("ec2", "describe_route_tables", "RouteTables", "NextToken", "NextToken", {}),
"security_groups": ("ec2", "describe_security_groups", "SecurityGroups", "NextToken", "NextToken", {}),
"nat_gateways": ("ec2", "describe_nat_gateways", "NatGateways", "NextToken", "NextToken", {}),
"internet_gateways": ("ec2", "describe_internet_gateways", "InternetGateways", "NextToken", "NextToken", {}),
"transit_gateways": ("ec2", "describe_transit_gateways", "TransitGateways", "NextToken", "NextToken", {}),
"vpc_peering": ("ec2", "describe_vpc_peering_connections", "VpcPeeringConnections", "NextToken", "NextToken", {}),
"vpc_endpoints": ("ec2", "describe_vpc_endpoints", "VpcEndpoints", "NextToken", "NextToken", {}),
"load_balancers": ("elbv2", "describe_load_balancers", "LoadBalancers", "Marker", "NextMarker", {}),
}
AWS_DATA = {
"rds_instances": ("rds", "describe_db_instances", "DBInstances", "Marker", "Marker", {}),
"dynamodb_tables": ("dynamodb", "list_tables", "TableNames", "ExclusiveStartTableName", "LastEvaluatedTableName", {}),
"s3_buckets": ("s3", "list_buckets", "Buckets", "ContinuationToken", "ContinuationToken", {}),
}
AWS_ORG_POLICY_TYPES = {"service_control_policies": "SERVICE_CONTROL_POLICY", "resource_control_policies": "RESOURCE_CONTROL_POLICY",
"tag_policies": "TAG_POLICY", "backup_policies": "BACKUP_POLICY"}
AWS_PILLARS = {"cost_optimizing": "cost", "security": "security", "fault_tolerance": "reliability",
"performance": "performance", "operational_excellence": "operations", "service_limits": "reliability"}
def _aws_csv(connection: Dict[str, Any], field: str) -> List[str]:
return [v.strip() for v in str(connection.get(field) or "").split(",") if v.strip()]
def _aws_session(connection: Dict[str, Any], account_id: Optional[str]):
"""A session on the customer's read-only role, or on the member account's when one is named."""
key = hashlib.sha256(json.dumps([connection, account_id], sort_keys=True).encode()).hexdigest()
cached = AWS_SESSIONS.get(key)
now = time.time()
if cached and cached[1] > now + 60:
return cached[0]
def assume(session, role_arn, external_id=None):
region = _aws_csv(connection, "regions")[0]
request = {"RoleArn": role_arn, "RoleSessionName": "kab-discovery", "DurationSeconds": 3600}
if external_id:
request["ExternalId"] = external_id
found = session.client("sts", region_name=region).assume_role(**request)["Credentials"]
return boto3.Session(aws_access_key_id=found["AccessKeyId"], aws_secret_access_key=found["SecretAccessKey"],
aws_session_token=found["SessionToken"])
base = boto3.Session(aws_access_key_id=connection["access_key_id"], aws_secret_access_key=connection["secret_access_key"])
session = assume(base, connection["role_arn"], connection.get("external_id"))
if account_id:
partition = str(connection["role_arn"]).split(":")[1]
session = assume(session, f"arn:{partition}:iam::{account_id}:role/{connection['member_role_name']}")
AWS_SESSIONS[key] = (session, now + 3000)
return session
def _aws_target(arguments: Dict[str, Any]):
"""(account, region) or a refusal. Nothing outside the connection's accounts and regions is ever reached."""
connection = get_connection()
regions = _aws_csv(connection, "regions")
region = arguments.get("region") or regions[0]
if region not in regions:
return None, None, _refuse("outside_scope", "region isn't in this connection's regions")
account = arguments.get("account_id")
if account and (account not in _aws_csv(connection, "account_ids") or not connection.get("member_role_name")):
return None, None, _refuse("outside_scope", "account_id isn't in this connection's accounts")
return account, region, None
async def _aws_call(arguments: Dict[str, Any], service: str, operation: str, region: Optional[str] = None, **request):
connection = get_connection()
account = arguments.get("account_id")
where = region or arguments.get("region") or _aws_csv(connection, "regions")[0]
def run():
# Adaptive mode slows the client down when AWS throttles it, on top of retrying.
retries = BotoConfig(retries={"max_attempts": 6, "mode": "adaptive"}, read_timeout=60, connect_timeout=10)
client = _aws_session(connection, account).client(service, region_name=where, config=retries)
return getattr(client, operation)(**request)
return await asyncio.to_thread(run)
async def _aws_kind(tool: str, arguments: Dict[str, Any], kinds: Dict[str, tuple], default: str) -> Dict[str, Any]:
kind = arguments.get("kind") or default
if kind not in kinds:
return _refuse("unknown_kind", "kind must be one of: " + ", ".join(kinds))
account, region, refused = _aws_target(arguments)
if refused:
return refused
service, operation, result_key, token_in, token_out, fixed = kinds[kind]
try:
upstream, skip = _read_page(arguments)
except (ValueError, KeyError, TypeError):
return _refuse("invalid_page_token", "page_token isn't one this server issued")
request = dict(fixed)
if upstream:
request[token_in] = upstream
try:
body = await _aws_call(arguments, service, operation, region, **request)
except Exception as error:
return _upstream_failed("aws", error)
extra = {"account_id": account, "region": region, "kind": kind}
return _page_result("aws", tool, arguments, body.get(result_key, []), upstream, body.get(token_out), skip, extra)
async def _aws_each(tool: str, arguments: Dict[str, Any], list_kind: tuple, read) -> Dict[str, Any]:
"""List one page of things, then read a detail for each. A detail that's absent or refused is reported."""
account, region, refused = _aws_target(arguments)
if refused:
return refused
service, operation, result_key, token_in, token_out, fixed = list_kind
try:
upstream, skip = _read_page(arguments)
except (ValueError, KeyError, TypeError):
return _refuse("invalid_page_token", "page_token isn't one this server issued")
request = dict(fixed)
if upstream:
request[token_in] = upstream
try:
body = await _aws_call(arguments, service, operation, region, **request)
except Exception as error:
return _upstream_failed("aws", error)
items = []
for entry in body.get(result_key, [])[skip:skip + _page_size(arguments)]:
try:
items.append(await read(entry, region))
except Exception as error:
items.append({"resource": entry, "unread": type(error).__name__})
consumed = skip + len(items)
more = consumed < len(body.get(result_key, []))
next_upstream = upstream if more else body.get(token_out)
result = _page_result("aws", tool, arguments, items, None, None, 0, {"account_id": account, "region": region})
if more or next_upstream:
result["data"]["next_page_token"] = _write_page(next_upstream, consumed if more else 0)
return result
def _aws_document(text: Any) -> Any:
"""A policy document, parsed when it arrives as a JSON string, so its principals can be scrubbed."""
try:
return json.loads(text) if isinstance(text, str) else text
except ValueError:
return text
async def aws_check_connection(arguments: Dict[str, Any]) -> Dict[str, Any]:
try:
who = await _aws_call(arguments, "sts", "get_caller_identity")
except Exception as error:
return {"ok": False, "message": _upstream_failed("aws", error)["error"]["message"]}
return {"ok": True, "account": who.get("Account"), "message": "Assumed the read-only role."}
async def aws_list_accounts(arguments: Dict[str, Any]) -> Dict[str, Any]:
kinds = {"accounts": ("organizations", "list_accounts", "Accounts", "NextToken", "NextToken", {}), "roots": ("organizations", "list_roots", "Roots", "NextToken", "NextToken", {})}
return await _aws_kind("aws_list_accounts", arguments, kinds, "accounts")
async def aws_list_resources(arguments: Dict[str, Any]) -> Dict[str, Any]:
resource_type = str(arguments.get("resource_type") or "")
if not re.fullmatch(r"AWS::[A-Za-z0-9]+::[A-Za-z0-9]+", resource_type):
return _refuse("invalid_argument", "resource_type must be a Config type, such as AWS::EC2::Instance")
kinds = {"config": ("config", "list_discovered_resources", "resourceIdentifiers", "nextToken", "nextToken", {"resourceType": resource_type})}
return await _aws_kind("aws_list_resources", {**arguments, "kind": "config"}, kinds, "config")
async def aws_query_resources(arguments: Dict[str, Any]) -> Dict[str, Any]:
query = str(arguments.get("query") or "").strip()
if not query.upper().startswith("SELECT ") or len(query) > 4000:
return _refuse("invalid_argument", "query must be a SELECT, up to 4000 characters")
aggregator = get_connection().get("aggregator_name")
if aggregator:
kinds = {"q": ("config", "select_aggregate_resource_config", "Results", "NextToken", "NextToken", {"Expression": query, "ConfigurationAggregatorName": aggregator})}
else:
kinds = {"q": ("config", "select_resource_config", "Results", "NextToken", "NextToken", {"Expression": query})}
result = await _aws_kind("aws_query_resources", {**arguments, "kind": "q"}, kinds, "q")
if result.get("success"):
result["data"]["items"] = [_scrub(_aws_document(i)) for i in result["data"]["items"]]
return result
async def aws_get_resource_metadata(arguments: Dict[str, Any]) -> Dict[str, Any]:
account, region, refused = _aws_target(arguments)
if refused:
return refused
try:
body = await _aws_call(arguments, "config", "get_resource_config_history", region, resourceType=arguments.get("resource_type"), resourceId=arguments.get("resource_id"), limit=1)
except Exception as error:
return _upstream_failed("aws", error)
items = body.get("configurationItems", [])
if not items:
return _refuse("not_found", "AWS Config holds no record of this resource")
return _one_result("aws", "aws_get_resource_metadata", arguments, _aws_document(items[0]))
async def aws_list_servers(arguments: Dict[str, Any]) -> Dict[str, Any]:
return await _aws_kind("aws_list_servers", arguments, AWS_SERVERS, "instances")
async def aws_list_workloads(arguments: Dict[str, Any]) -> Dict[str, Any]:
return await _aws_kind("aws_list_workloads", arguments, AWS_WORKLOADS, "lambda_functions")
async def aws_list_kubernetes_clusters(arguments: Dict[str, Any]) -> Dict[str, Any]:
async def read(name, region):
cluster = (await _aws_call(arguments, "eks", "describe_cluster", region, name=name))["cluster"]
groups = await _aws_call(arguments, "eks", "list_nodegroups", region, clusterName=name)
return {**cluster, "nodegroups": groups.get("nodegroups", [])}
return await _aws_each("aws_list_kubernetes_clusters", arguments, ("eks", "list_clusters", "clusters", "nextToken", "nextToken", {}), read)
async def aws_get_network_topology(arguments: Dict[str, Any]) -> Dict[str, Any]:
return await _aws_kind("aws_get_network_topology", arguments, AWS_NETWORK, "vpcs")
async def aws_list_data_stores(arguments: Dict[str, Any]) -> Dict[str, Any]:
return await _aws_kind("aws_list_data_stores", arguments, AWS_DATA, "rds_instances")
async def aws_list_organization_policies(arguments: Dict[str, Any]) -> Dict[str, Any]:
kind = arguments.get("kind") or "service_control_policies"
if kind == "config_rule_compliance":
kinds = {kind: ("config", "describe_compliance_by_config_rule", "ComplianceByConfigRules", "NextToken", "NextToken", {})}
return await _aws_kind("aws_list_organization_policies", arguments, kinds, kind)
if kind not in AWS_ORG_POLICY_TYPES:
return _refuse("unknown_kind", "kind must be one of: " + ", ".join(AWS_ORG_POLICY_TYPES) + ", config_rule_compliance")
async def read(entry, region):
policy = (await _aws_call(arguments, "organizations", "describe_policy", region, PolicyId=entry["Id"]))["Policy"]
targets = await _aws_call(arguments, "organizations", "list_targets_for_policy", region, PolicyId=entry["Id"])
return {"policy_id": entry["Id"], "name": entry.get("Name"), "type": entry.get("Type"), "layer": "organisation", "aws_managed": entry.get("AwsManaged"), "attached_to": targets.get("Targets", []), "document": _aws_document(policy.get("Content"))}
return await _aws_each("aws_list_organization_policies", arguments, ("organizations", "list_policies", "Policies", "NextToken", "NextToken", {"Filter": AWS_ORG_POLICY_TYPES[kind]}), read)
async def aws_list_resource_policies(arguments: Dict[str, Any]) -> Dict[str, Any]:
kind = arguments.get("kind") or "s3_buckets"
def shaped(name, document, resource_type):
return {"name": name, "type": resource_type, "layer": "resource", "document": _aws_document(document)}
async def bucket(entry, region):
body = await _aws_call(arguments, "s3", "get_bucket_policy", region, Bucket=entry["Name"])
return shaped(entry["Name"], body.get("Policy"), "s3_bucket")
async def key(entry, region):
body = await _aws_call(arguments, "kms", "get_key_policy", region, KeyId=entry["KeyId"], PolicyName="default")
return shaped(entry["KeyId"], body.get("Policy"), "kms_key")
async def function(entry, region):
body = await _aws_call(arguments, "lambda", "get_policy", region, FunctionName=entry["FunctionName"])
return shaped(entry["FunctionName"], body.get("Policy"), "lambda_function")
async def queue(url, region):
body = await _aws_call(arguments, "sqs", "get_queue_attributes", region, QueueUrl=url, AttributeNames=["Policy"])
return shaped(url, body.get("Attributes", {}).get("Policy"), "sqs_queue")
async def topic(entry, region):
body = await _aws_call(arguments, "sns", "get_topic_attributes", region, TopicArn=entry["TopicArn"])
return shaped(entry["TopicArn"], body.get("Attributes", {}).get("Policy"), "sns_topic")
async def role(entry, region):
return shaped(entry["RoleName"], entry.get("AssumeRolePolicyDocument"), "iam_role_trust")
kinds = {"s3_buckets": (("s3", "list_buckets", "Buckets", "ContinuationToken", "ContinuationToken", {}), bucket),
"kms_keys": (("kms", "list_keys", "Keys", "Marker", "NextMarker", {}), key),
"lambda_functions": (("lambda", "list_functions", "Functions", "Marker", "NextMarker", {}), function),
"sqs_queues": (("sqs", "list_queues", "QueueUrls", "NextToken", "NextToken", {}), queue),
"sns_topics": (("sns", "list_topics", "Topics", "NextToken", "NextToken", {}), topic),
"iam_role_trust": (("iam", "list_roles", "Roles", "Marker", "Marker", {}), role)}
if kind not in kinds:
return _refuse("unknown_kind", "kind must be one of: " + ", ".join(kinds))
return await _aws_each("aws_list_resource_policies", arguments, kinds[kind][0], kinds[kind][1])
async def aws_list_recommendations(arguments: Dict[str, Any]) -> Dict[str, Any]:
kind = arguments.get("kind") or "trusted_advisor"
if kind == "well_architected_reviews":
async def read(entry, region):
lenses = await _aws_call(arguments, "wellarchitected", "list_lens_reviews", region, WorkloadId=entry["WorkloadId"])
return {"workload": entry, "lens_reviews": lenses.get("LensReviewSummaries", [])}
return await _aws_each("aws_list_recommendations", arguments, ("wellarchitected", "list_workloads", "WorkloadSummaries", "NextToken", "NextToken", {}), read)
if kind != "trusted_advisor":
return _refuse("unknown_kind", "kind must be trusted_advisor or well_architected_reviews")
async def read(entry, region):
pillars = entry.get("pillars") or []
return {**entry, "pillar": [AWS_PILLARS.get(p, "other") for p in pillars], "provider_category": pillars}
return await _aws_each("aws_list_recommendations", arguments, ("trustedadvisor", "list_recommendations", "recommendationSummaries", "nextToken", "nextToken", {}), read)
async def verify_connection(arguments: Dict[str, Any]) -> Dict[str, Any]:
"""Check that the connection sent with this request works against the upstream. Makes no change."""
connection = get_connection()
if not connection:
return {"ok": False, "message": "No connection was sent with this request."}
missing = mcp_connection.missing_fields(connection, CONNECTION_SCHEMA)
if missing:
return {"ok": False, "message": "The connection is missing: " + ", ".join(missing)}
return await aws_check_connection(arguments)
# Tool function registry
TOOL_FUNCTIONS = {
'aws_check_connection': aws_check_connection,
'aws_list_accounts': aws_list_accounts,
'aws_list_resources': aws_list_resources,
'aws_query_resources': aws_query_resources,
'aws_get_resource_metadata': aws_get_resource_metadata,
'aws_list_servers': aws_list_servers,
'aws_list_workloads': aws_list_workloads,
'aws_list_kubernetes_clusters': aws_list_kubernetes_clusters,
'aws_get_network_topology': aws_get_network_topology,
'aws_list_data_stores': aws_list_data_stores,
'aws_list_organization_policies': aws_list_organization_policies,
'aws_list_resource_policies': aws_list_resource_policies,
'aws_list_recommendations': aws_list_recommendations,
'verify_connection': verify_connection,
}
# ============================================
# Resource Implementations
# ============================================
# Resource read handlers by uri
RESOURCE_HANDLERS = {
}
# ============================================
# FastAPI Application
# ============================================
@asynccontextmanager
async def lifespan(app: FastAPI):
"""Application lifespan manager"""
logger.info(
"server_starting",
port=config.SERVER_PORT,
auth_required=config.REQUIRE_AUTH,
tools=len(TOOLS_REGISTRY),
resources=len(RESOURCES_REGISTRY),
prompts=len(PROMPTS_REGISTRY),
template_version=TEMPLATE_VERSION,
)
# Validate configuration
if config.REQUIRE_AUTH and not config.MCP_REGISTRATION_API_KEY:
logger.error(
"registration_key_not_set",
hint="Set MCP_REGISTRATION_API_KEY before starting the server",
)
yield
await _close_all_clients()
logger.info("server_stopping")
app = FastAPI(
title=config.SERVER_NAME,
version=config.SERVER_VERSION,
description="Read-only discovery of a customer's AWS estate - servers, workloads, Kubernetes clusters, networks, data stores, policies and Trusted Advisor findings.",
lifespan=lifespan
)
# FastAPI's default answer to an invalid request repeats it, and the request can carry the tenant's connection
mcp_connection.install_invalid_request_handler(
app, lambda fields: logger.warning("invalid_request", fields=sanitize_log(fields))
)
# ============================================
# Health Check Endpoint
# ============================================
@app.get("/health")
async def health_check():
"""
Health check endpoint (no authentication required).
Used by:
- MCP Gateway health monitoring
- Kubernetes liveness/readiness probes
- Load balancers
"""
# Answers without a key, so it says only what a probe needs: not how authentication is set up
return {
"status": "healthy" if config.MCP_REGISTRATION_API_KEY else "degraded",
"server_id": config.SERVER_ID,
"server_name": config.SERVER_NAME,
"version": config.SERVER_VERSION,
"timestamp": datetime.now(timezone.utc).isoformat(),
}
# ============================================
# MCP Authentication Info Endpoint
# ============================================
@app.get("/.well-known/mcp-auth")
async def mcp_auth_info():
"""
MCP Authentication Information Endpoint (PUBLIC - no auth required).
Exposes authentication requirements for Gateway registration.
"""
return {
"auth_required": config.REQUIRE_AUTH,
"auth_type": "bearer",
"description": "Include 'Authorization: Bearer <MCP_REGISTRATION_API_KEY>' header",
"env_var": "MCP_REGISTRATION_API_KEY"
}
# ============================================
# MCP Protocol /messages Endpoint
# ============================================
class _MethodNotFound(Exception):
"""The JSON-RPC method is not one this server answers."""
async def dispatch_mcp_request(request: MCPRequest) -> MCPResponse:
"""Run one JSON-RPC request; every outcome, errors included, is an MCPResponse."""
# The tenant's connection, bound for this request only so tools reach the right account
connection_token = mcp_connection.bind(mcp_connection.from_params(request.params))
try:
if CONNECTION_SCHEMA is not None and request.method in CONNECTION_METHODS:
check_connection(mcp_connection.active())
if request.method == "initialize":
result = {
"protocolVersion": "1.0.0",
"serverInfo": {
"name": config.SERVER_NAME,
"version": config.SERVER_VERSION,
"server_id": config.SERVER_ID
},
"capabilities": {
"tools": True,
"resources": False,
"prompts": False
},
"authentication": {
"type": "bearer",
"description": "Requires MCP_REGISTRATION_API_KEY in Authorization header"
}
}
if CONNECTION_SCHEMA is not None:
result["_meta"] = mcp_connection.schema_meta(CONNECTION_SCHEMA, verify_tool=VERIFY_TOOL)
return MCPResponse(id=request.id, result=result)
elif request.method == "tools/list":
return MCPResponse(id=request.id, result={"tools": TOOLS_REGISTRY})
elif request.method == "tools/call":
tool_name = request.params.get("name")
arguments = request.params.get("arguments")
if arguments is None:
arguments = {}
if not tool_name:
raise ValueError("Missing 'name' in params")
if not isinstance(tool_name, str):
raise ValueError("'name' must be a string")
if not isinstance(arguments, dict):
raise ValueError("'arguments' must be an object")
structlog.contextvars.bind_contextvars(tool=sanitize_log(tool_name))
tool_func = TOOL_FUNCTIONS.get(tool_name)
if not tool_func:
raise ValueError(f"Unknown tool: {tool_name}")
result = await tool_func(arguments)
if isinstance(result, dict) and result.get("success") is False and isinstance(result.get("error"), str):
# A failed call's message often quotes the upstream URL, which may carry a key
result = {**result, "error": mcp_connection.scrub(result["error"])}
reply = {
"content": [
{
"type": "text",
"text": json.dumps(result) if isinstance(result, dict) else str(result)
}
]
}
if isinstance(result, dict) and result.get("success") is False:
# A tool that reports failure is an error to the caller too (MCP: isError), not a normal result
reply["isError"] = True
return MCPResponse(id=request.id, result=reply)
elif request.method == "resources/list":
return MCPResponse(id=request.id, result={"resources": RESOURCES_REGISTRY})
elif request.method == "resources/read":
uri = request.params.get("uri")
if not uri:
raise ValueError("Missing 'uri' in params")
structlog.contextvars.bind_contextvars(resource=sanitize_log(uri))
handler = RESOURCE_HANDLERS.get(uri)
if not handler:
raise ValueError(f"Unknown resource: {uri}")
content = await handler(uri)
mime_type = next((r["mimeType"] for r in RESOURCES_REGISTRY if r["uri"] == uri), "text/plain")
text = content if isinstance(content, str) else json.dumps(content)
return MCPResponse(id=request.id, result={
"contents": [{"uri": uri, "mimeType": mime_type, "text": text}]
})
elif request.method == "prompts/list":
return MCPResponse(id=request.id, result={"prompts": PROMPTS_REGISTRY})
elif request.method == "prompts/get":
prompt_name = request.params.get("name")
prompt = PROMPT_TEMPLATES.get(prompt_name) if isinstance(prompt_name, str) else None
if prompt is None:
raise ValueError(f"Unknown prompt: {prompt_name}")
structlog.contextvars.bind_contextvars(prompt=sanitize_log(prompt_name))
supplied = request.params.get("arguments") or {}
values = {a["name"]: a["default"] for a in prompt["arguments"] if a["default"] is not None}
values.update(supplied)
missing = [a["name"] for a in prompt["arguments"] if a["required"] and a["name"] not in values]
if missing:
raise ValueError("Missing prompt argument(s): " + ", ".join(missing))
return MCPResponse(id=request.id, result={
"description": prompt["description"],
"messages": [{
"role": "user",
"content": {"type": "text", "text": render_prompt(prompt["template"], values)},
}],
})
else:
raise _MethodNotFound(f"Unknown method: {request.method}")
except _MethodNotFound as e:
logger.warning("mcp_method_not_found", error=sanitize_log(mcp_connection.redact(e)))
return MCPResponse(id=request.id, error={"code": -32601, "message": mcp_connection.redact(e)})
except NotImplementedError as e:
message = mcp_connection.redact(e)
logger.warning("tool_not_implemented", error=sanitize_log(message))
return MCPResponse(id=request.id, error={"code": -32601, "message": message})
except ValueError as e:
message = mcp_connection.redact(e)
logger.warning("mcp_client_error", error=sanitize_log(message))
return MCPResponse(id=request.id, error={"code": -32602, "message": message})
except Exception as e:
# The message of an unexpected error is not shown to the caller and is redacted before it is logged;
# a tool may have put the connection's credential in it.
logger.error(
"mcp_unexpected_error",
error_type=type(e).__name__,
error=sanitize_log(mcp_connection.redact(e)),
traceback=sanitize_log(mcp_connection.redact(traceback.format_exc())),
)
return MCPResponse(id=request.id, error={"code": -32603, "message": "Internal server error"})
finally:
mcp_connection.unbind(connection_token)
@app.post("/messages")
async def handle_mcp_message(
request: MCPRequest,
http_request: Request,
authenticated: bool = Depends(verify_mcp_registration_key)
) -> MCPResponse:
"""
MCP protocol message handler (JSON-RPC 2.0 over HTTP).
This is the STANDARD MCP endpoint for Gateway integration.
Requires Bearer token authentication (MCP_REGISTRATION_API_KEY).
Supported Methods:
- initialize: Protocol handshake and capability exchange
- tools/list: Discover available tools
- tools/call: Execute a specific tool
- resources/list: List available resources
- prompts/list: List available prompts
"""
started = time.monotonic()
structlog.contextvars.clear_contextvars()
structlog.contextvars.bind_contextvars(
request_id=sanitize_log(http_request.headers.get("x-request-id") or str(request.id)),
method=sanitize_log(request.method),
)
try:
response = await dispatch_mcp_request(request)
completed = {}
if CONNECTION_SCHEMA is not None and request.method in CONNECTION_METHODS:
# Which connection served the call: a short id, never the credential
completed["connection"] = mcp_connection.fingerprint(mcp_connection.from_params(request.params))
logger.info(
"mcp_request_completed",
outcome="error" if response.error else ("tool_error" if (response.result or {}).get("isError") else "ok"),
error_code=(response.error or {}).get("code"),
duration_ms=round((time.monotonic() - started) * 1000, 1),
**completed,
)
return response
finally:
structlog.contextvars.clear_contextvars()
# ============================================
# Entry Point
# ============================================
if __name__ == "__main__":
import uvicorn
uvicorn.run(
"server:app",
host="0.0.0.0",
port=config.SERVER_PORT,
log_level=logging.getLevelName(getattr(logging, config.LOG_LEVEL, logging.INFO)).lower(),
log_config=None, # configure_logging() owns the format
access_log=True,
reload=False # Set to True for development
)