nanobot/nanobot/agent/tools/mcp.py

"""MCP client: connects to MCP servers and wraps their tools as native nanobot tools."""

import asyncio
import os
import re
import shutil
import urllib.parse
from collections.abc import Awaitable, Callable
from contextlib import AsyncExitStack, suppress
from typing import Any, Mapping
from weakref import WeakKeyDictionary

import httpx
from loguru import logger

from nanobot.agent.tools.base import Tool
from nanobot.agent.tools.registry import ToolRegistry
from nanobot.bus.events import (
    INBOUND_META_RUNTIME_CONTROL,
    RUNTIME_CONTROL_ACK,
    RUNTIME_CONTROL_MCP_RELOAD,
    InboundMessage,
)
from nanobot.security.network import validate_url_target

# Transient connection errors that warrant a single retry.
# These typically happen when an MCP server restarts or a network
# connection is interrupted between calls.
_TRANSIENT_EXC_NAMES: frozenset[str] = frozenset((
    "ClosedResourceError",
    "BrokenResourceError",
    "EndOfStream",
    "BrokenPipeError",
    "ConnectionResetError",
    "ConnectionRefusedError",
    "ConnectionAbortedError",
    "ConnectionError",
))

_WINDOWS_SHELL_LAUNCHERS: frozenset[str] = frozenset(("npx", "npm", "pnpm", "yarn", "bunx"))

# Characters allowed in tool names by model providers (Anthropic, OpenAI, etc.).
# Replace anything outside [a-zA-Z0-9_-] with underscore and collapse runs.
_SANITIZE_RE = re.compile(r"_+")
_RELOAD_LOCKS: WeakKeyDictionary[Any, asyncio.Lock] = WeakKeyDictionary()
_ReconnectCallback = Callable[[str, str, Tool], Awaitable[Tool | None]]


def _is_malformed_mcp_progress_notification(message: Any) -> bool:
    root = getattr(getattr(message, "message", None), "root", None)
    if getattr(root, "method", None) != "notifications/progress":
        return False

    params = getattr(root, "params", None)
    return not isinstance(params, Mapping) or "progressToken" not in params


class _MalformedProgressNotificationFilter:
    def __init__(self, read_stream: Any, server_name: str) -> None:
        self._read_stream = read_stream
        self._server_name = server_name
        self._iterator: Any | None = None

    async def __aenter__(self) -> "_MalformedProgressNotificationFilter":
        await self._read_stream.__aenter__()
        return self

    async def __aexit__(self, exc_type: Any, exc: Any, tb: Any) -> Any:
        return await self._read_stream.__aexit__(exc_type, exc, tb)

    def __aiter__(self) -> "_MalformedProgressNotificationFilter":
        self._iterator = self._read_stream.__aiter__()
        return self

    async def __anext__(self) -> Any:
        if self._iterator is None:
            self._iterator = self._read_stream.__aiter__()

        while True:
            message = await self._iterator.__anext__()
            if _is_malformed_mcp_progress_notification(message):
                logger.debug(
                    "MCP server '{}': dropped progress notification without progressToken",
                    self._server_name,
                )
                continue
            return message

    async def aclose(self) -> None:
        close = getattr(self._read_stream, "aclose", None)
        if close is not None:
            await close()


def _filter_malformed_mcp_progress_notifications(read_stream: Any, server_name: str) -> Any:
    if not all(hasattr(read_stream, name) for name in ("__aenter__", "__aexit__", "__aiter__")):
        return read_stream
    return _MalformedProgressNotificationFilter(read_stream, server_name)


def _sanitize_name(name: str) -> str:
    """Sanitize an MCP-derived name for model API compatibility."""
    return _SANITIZE_RE.sub("_", re.sub(r"[^a-zA-Z0-9_-]", "_", name))


def _is_transient(exc: BaseException) -> bool:
    """Check if an exception looks like a transient connection error."""
    return type(exc).__name__ in _TRANSIENT_EXC_NAMES


def _is_session_terminated(exc: BaseException) -> bool:
    """Return True when the MCP SDK reports a dead client session."""
    messages = [str(exc)]
    error = getattr(exc, "error", None)
    if error is not None:
        messages.append(str(getattr(error, "message", "")))
    return any(
        marker in message.lower()
        for marker in ("session terminated", "connection closed")
        for message in messages
    )


async def _probe_http_url(url: str, timeout: float = 3.0) -> bool:
    """Quick TCP probe to check if an HTTP MCP server is reachable.

    Avoids entering ``streamable_http_client`` / ``sse_client`` when the port is
    closed — those transports use anyio task groups whose cleanup can raise
    ``RuntimeError`` / ``ExceptionGroup`` that escape the caller's try/except
    and crash the event loop.
    """
    parsed = urllib.parse.urlparse(url)
    host = parsed.hostname or "127.0.0.1"
    port = parsed.port
    if not port:
        port = 443 if parsed.scheme == "https" else 80
    try:
        reader, writer = await asyncio.wait_for(
            asyncio.open_connection(host, port),
            timeout=timeout,
        )
        writer.close()
        with suppress(OSError, asyncio.TimeoutError):
            await asyncio.wait_for(writer.wait_closed(), timeout=0.2)
        return True
    except (OSError, asyncio.TimeoutError):
        return False


async def _validate_mcp_request_url(request: httpx.Request) -> None:
    """Validate each outgoing MCP HTTP request, including redirect targets."""
    ok, error = validate_url_target(str(request.url))
    if not ok:
        raise httpx.RequestError(
            f"Blocked unsafe MCP URL {request.url} ({error})",
            request=request,
        )


def _windows_command_basename(command: str) -> str:
    """Return the lowercase basename for a Windows command or path."""
    return command.replace("\\", "/").rsplit("/", maxsplit=1)[-1].lower()


def _normalize_windows_stdio_command(
    command: str,
    args: list[str] | None,
    env: dict[str, str] | None,
) -> tuple[str, list[str], dict[str, str] | None]:
    """Wrap Windows shell launchers so MCP stdio servers start reliably."""
    normalized_args = list(args or [])
    if os.name != "nt":
        return command, normalized_args, env

    basename = _windows_command_basename(command)
    if basename in {"cmd", "cmd.exe", "powershell", "powershell.exe", "pwsh", "pwsh.exe"}:
        return command, normalized_args, env

    if basename.endswith((".exe", ".com")):
        return command, normalized_args, env

    resolved = shutil.which(command, path=(env or {}).get("PATH")) or command
    resolved_basename = _windows_command_basename(resolved)
    should_wrap = (
        basename in _WINDOWS_SHELL_LAUNCHERS
        or basename.endswith((".cmd", ".bat"))
        or resolved_basename.endswith((".cmd", ".bat"))
    )
    if not should_wrap:
        return command, normalized_args, env

    comspec = (env or {}).get("COMSPEC") or os.environ.get("COMSPEC") or "cmd.exe"
    return comspec, ["/d", "/c", command, *normalized_args], env


def _extract_nullable_branch(options: Any) -> tuple[dict[str, Any], bool] | None:
    """Return the single non-null branch for nullable unions."""
    if not isinstance(options, list):
        return None

    non_null: list[dict[str, Any]] = []
    saw_null = False
    for option in options:
        if not isinstance(option, dict):
            return None
        if option.get("type") == "null":
            saw_null = True
            continue
        non_null.append(option)

    if saw_null and len(non_null) == 1:
        return non_null[0], True
    return None


def _normalize_schema_for_openai(schema: Any) -> dict[str, Any]:
    """Normalize only nullable JSON Schema patterns for tool definitions."""
    if not isinstance(schema, dict):
        return {"type": "object", "properties": {}}

    normalized = dict(schema)

    raw_type = normalized.get("type")
    if isinstance(raw_type, list):
        non_null = [item for item in raw_type if item != "null"]
        if "null" in raw_type and len(non_null) == 1:
            normalized["type"] = non_null[0]
            normalized["nullable"] = True

    for key in ("oneOf", "anyOf"):
        nullable_branch = _extract_nullable_branch(normalized.get(key))
        if nullable_branch is not None:
            branch, _ = nullable_branch
            merged = {k: v for k, v in normalized.items() if k != key}
            merged.update(branch)
            normalized = merged
            normalized["nullable"] = True
            break

    if "properties" in normalized and isinstance(normalized["properties"], dict):
        normalized["properties"] = {
            name: _normalize_schema_for_openai(prop) if isinstance(prop, dict) else prop
            for name, prop in normalized["properties"].items()
        }

    if "items" in normalized and isinstance(normalized["items"], dict):
        normalized["items"] = _normalize_schema_for_openai(normalized["items"])

    if normalized.get("type") != "object":
        return normalized

    normalized.setdefault("properties", {})
    normalized.setdefault("required", [])
    return normalized


class _MCPWrapperBase(Tool):
    """Common reconnect handling for wrappers bound to one MCP server session."""

    _plugin_discoverable = False

    def _set_mcp_connection(self, session: Any, server_name: str) -> None:
        self._session = session
        self._server_name = server_name
        self._reconnect: _ReconnectCallback | None = None

    def set_reconnect_handler(self, reconnect: _ReconnectCallback) -> None:
        self._reconnect = reconnect

    async def _refresh_session_after_termination(
        self,
        exc: BaseException,
        already_refreshed: bool,
        capability_kind: str,
    ) -> bool:
        if already_refreshed or not _is_session_terminated(exc) or self._reconnect is None:
            return False
        logger.warning(
            "MCP {} '{}' session terminated; reconnecting server '{}' before retry",
            capability_kind,
            self._name,
            self._server_name,
        )
        refreshed_tool = await self._reconnect(self._server_name, self._name, self)
        refreshed_session = getattr(refreshed_tool, "_session", None)
        if refreshed_session is None:
            logger.warning(
                "MCP {} '{}' could not refresh session for server '{}'",
                capability_kind,
                self._name,
                self._server_name,
            )
            return False
        self._session = refreshed_session
        return True


class MCPToolWrapper(_MCPWrapperBase):
    """Wraps a single MCP server tool as a nanobot Tool."""

    _plugin_discoverable = False

    def __init__(self, session, server_name: str, tool_def, tool_timeout: int = 30):
        self._set_mcp_connection(session, server_name)
        self._original_name = tool_def.name
        self._name = _sanitize_name(f"mcp_{server_name}_{tool_def.name}")
        self._description = tool_def.description or tool_def.name
        raw_schema = tool_def.inputSchema or {"type": "object", "properties": {}}
        self._parameters = _normalize_schema_for_openai(raw_schema)
        self._tool_timeout = tool_timeout

    @property
    def name(self) -> str:
        return self._name

    @property
    def description(self) -> str:
        return self._description

    @property
    def parameters(self) -> dict[str, Any]:
        return self._parameters

    async def execute(self, **kwargs: Any) -> str:
        from mcp import types

        retried_transient = False
        refreshed_session = False
        while True:
            try:
                result = await asyncio.wait_for(
                    self._session.call_tool(self._original_name, arguments=kwargs),
                    timeout=self._tool_timeout,
                )
            except asyncio.TimeoutError:
                logger.warning(
                    "MCP tool '{}' timed out after {}s", self._name, self._tool_timeout
                )
                return f"(MCP tool call timed out after {self._tool_timeout}s)"
            except asyncio.CancelledError:
                # MCP SDK's anyio cancel scopes can leak CancelledError on timeout/failure.
                # Re-raise only if our task was externally cancelled (e.g. /stop).
                task = asyncio.current_task()
                if task is not None and task.cancelling() > 0:
                    raise
                logger.warning("MCP tool '{}' was cancelled by server/SDK", self._name)
                return "(MCP tool call was cancelled)"
            except Exception as exc:
                if await self._refresh_session_after_termination(
                    exc,
                    refreshed_session,
                    "tool",
                ):
                    refreshed_session = True
                    continue
                if _is_transient(exc):
                    if not retried_transient:
                        retried_transient = True
                        logger.warning(
                            "MCP tool '{}' hit transient error ({}), retrying once...",
                            self._name,
                            type(exc).__name__,
                        )
                        await asyncio.sleep(1)  # Brief backoff before retry
                        continue
                    # Second transient failure — give up with retry-specific message
                    logger.exception(
                        "MCP tool '{}' failed after retry: {}",
                        self._name,
                        type(exc).__name__,
                    )
                    return f"(MCP tool call failed after retry: {type(exc).__name__})"
                logger.exception(
                    "MCP tool '{}' failed: {}: {}",
                    self._name,
                    type(exc).__name__,
                    exc,
                )
                return f"(MCP tool call failed: {type(exc).__name__})"
            else:
                # Success — extract result
                parts = []
                for block in result.content:
                    if isinstance(block, types.TextContent):
                        parts.append(block.text)
                    else:
                        parts.append(str(block))
                return "\n".join(parts) or "(no output)"

        return "(MCP tool call failed)"  # Unreachable, but satisfies type checkers


class MCPResourceWrapper(_MCPWrapperBase):
    """Wraps an MCP resource URI as a read-only nanobot Tool."""

    _plugin_discoverable = False

    def __init__(self, session, server_name: str, resource_def, resource_timeout: int = 30):
        self._set_mcp_connection(session, server_name)
        self._uri = resource_def.uri
        self._name = _sanitize_name(f"mcp_{server_name}_resource_{resource_def.name}")
        desc = resource_def.description or resource_def.name
        self._description = f"[MCP Resource] {desc}\nURI: {self._uri}"
        self._parameters: dict[str, Any] = {
            "type": "object",
            "properties": {},
            "required": [],
        }
        self._resource_timeout = resource_timeout

    @property
    def name(self) -> str:
        return self._name

    @property
    def description(self) -> str:
        return self._description

    @property
    def parameters(self) -> dict[str, Any]:
        return self._parameters

    @property
    def read_only(self) -> bool:
        return True

    async def execute(self, **kwargs: Any) -> str:
        from mcp import types

        retried_transient = False
        refreshed_session = False
        while True:
            try:
                result = await asyncio.wait_for(
                    self._session.read_resource(self._uri),
                    timeout=self._resource_timeout,
                )
            except asyncio.TimeoutError:
                logger.warning(
                    "MCP resource '{}' timed out after {}s", self._name, self._resource_timeout
                )
                return f"(MCP resource read timed out after {self._resource_timeout}s)"
            except asyncio.CancelledError:
                task = asyncio.current_task()
                if task is not None and task.cancelling() > 0:
                    raise
                logger.warning("MCP resource '{}' was cancelled by server/SDK", self._name)
                return "(MCP resource read was cancelled)"
            except Exception as exc:
                if await self._refresh_session_after_termination(
                    exc,
                    refreshed_session,
                    "resource",
                ):
                    refreshed_session = True
                    continue
                if _is_transient(exc):
                    if not retried_transient:
                        retried_transient = True
                        logger.warning(
                            "MCP resource '{}' hit transient error ({}), retrying once...",
                            self._name,
                            type(exc).__name__,
                        )
                        await asyncio.sleep(1)
                        continue
                    logger.exception(
                        "MCP resource '{}' failed after retry: {}",
                        self._name,
                        type(exc).__name__,
                    )
                    return f"(MCP resource read failed after retry: {type(exc).__name__})"
                logger.exception(
                    "MCP resource '{}' failed: {}: {}",
                    self._name,
                    type(exc).__name__,
                    exc,
                )
                return f"(MCP resource read failed: {type(exc).__name__})"
            else:
                parts: list[str] = []
                for block in result.contents:
                    if isinstance(block, types.TextResourceContents):
                        parts.append(block.text)
                    elif isinstance(block, types.BlobResourceContents):
                        parts.append(f"[Binary resource: {len(block.blob)} bytes]")
                    else:
                        parts.append(str(block))
                return "\n".join(parts) or "(no output)"

        return "(MCP resource read failed)"  # Unreachable


class MCPPromptWrapper(_MCPWrapperBase):
    """Wraps an MCP prompt as a read-only nanobot Tool."""

    _plugin_discoverable = False

    def __init__(self, session, server_name: str, prompt_def, prompt_timeout: int = 30):
        self._set_mcp_connection(session, server_name)
        self._prompt_name = prompt_def.name
        self._name = _sanitize_name(f"mcp_{server_name}_prompt_{prompt_def.name}")
        desc = prompt_def.description or prompt_def.name
        self._description = (
            f"[MCP Prompt] {desc}\n"
            "Returns a filled prompt template that can be used as a workflow guide."
        )
        self._prompt_timeout = prompt_timeout

        # Build parameters from prompt arguments
        properties: dict[str, Any] = {}
        required: list[str] = []
        for arg in prompt_def.arguments or []:
            prop: dict[str, Any] = {"type": "string"}
            if getattr(arg, "description", None):
                prop["description"] = arg.description
            properties[arg.name] = prop
            if arg.required:
                required.append(arg.name)
        self._parameters: dict[str, Any] = {
            "type": "object",
            "properties": properties,
            "required": required,
        }

    @property
    def name(self) -> str:
        return self._name

    @property
    def description(self) -> str:
        return self._description

    @property
    def parameters(self) -> dict[str, Any]:
        return self._parameters

    @property
    def read_only(self) -> bool:
        return True

    async def execute(self, **kwargs: Any) -> str:
        from mcp import types
        from mcp.shared.exceptions import McpError

        retried_transient = False
        refreshed_session = False
        while True:
            try:
                result = await asyncio.wait_for(
                    self._session.get_prompt(self._prompt_name, arguments=kwargs),
                    timeout=self._prompt_timeout,
                )
            except asyncio.TimeoutError:
                logger.warning(
                    "MCP prompt '{}' timed out after {}s", self._name, self._prompt_timeout
                )
                return f"(MCP prompt call timed out after {self._prompt_timeout}s)"
            except asyncio.CancelledError:
                task = asyncio.current_task()
                if task is not None and task.cancelling() > 0:
                    raise
                logger.warning("MCP prompt '{}' was cancelled by server/SDK", self._name)
                return "(MCP prompt call was cancelled)"
            except McpError as exc:
                if await self._refresh_session_after_termination(
                    exc,
                    refreshed_session,
                    "prompt",
                ):
                    refreshed_session = True
                    continue
                logger.exception(
                    "MCP prompt '{}' failed: code={} message={}",
                    self._name,
                    exc.error.code,
                    exc.error.message,
                )
                return f"(MCP prompt call failed: {exc.error.message} [code {exc.error.code}])"
            except Exception as exc:
                if await self._refresh_session_after_termination(
                    exc,
                    refreshed_session,
                    "prompt",
                ):
                    refreshed_session = True
                    continue
                if _is_transient(exc):
                    if not retried_transient:
                        retried_transient = True
                        logger.warning(
                            "MCP prompt '{}' hit transient error ({}), retrying once...",
                            self._name,
                            type(exc).__name__,
                        )
                        await asyncio.sleep(1)
                        continue
                    logger.exception(
                        "MCP prompt '{}' failed after retry: {}",
                        self._name,
                        type(exc).__name__,
                    )
                    return f"(MCP prompt call failed after retry: {type(exc).__name__})"
                logger.exception(
                    "MCP prompt '{}' failed: {}: {}",
                    self._name,
                    type(exc).__name__,
                    exc,
                )
                return f"(MCP prompt call failed: {type(exc).__name__})"
            else:
                parts: list[str] = []
                for message in result.messages:
                    content = message.content
                    if isinstance(content, types.TextContent):
                        parts.append(content.text)
                    elif isinstance(content, list):
                        for block in content:
                            if isinstance(block, types.TextContent):
                                parts.append(block.text)
                            else:
                                parts.append(str(block))
                    else:
                        parts.append(str(content))
                return "\n".join(parts) or "(no output)"

        return "(MCP prompt call failed)"  # Unreachable


async def connect_mcp_servers(
    mcp_servers: dict, registry: ToolRegistry
) -> dict[str, AsyncExitStack]:
    """Connect to configured MCP servers and register their tools, resources, prompts.

    Returns a dict mapping server name -> its dedicated AsyncExitStack.
    Each server gets its own stack to prevent cancel scope conflicts
    when multiple MCP servers are configured.
    """
    from mcp import ClientSession, StdioServerParameters
    from mcp.client.sse import sse_client
    from mcp.client.stdio import stdio_client
    from mcp.client.streamable_http import streamable_http_client

    async def connect_single_server(name: str, cfg) -> tuple[str, AsyncExitStack | None]:
        server_stack = AsyncExitStack()
        await server_stack.__aenter__()

        try:
            transport_type = cfg.type
            if not transport_type:
                if cfg.command:
                    transport_type = "stdio"
                elif cfg.url:
                    transport_type = (
                        "sse" if cfg.url.rstrip("/").endswith("/sse") else "streamableHttp"
                    )
                else:
                    logger.warning("MCP server '{}': no command or url configured, skipping", name)
                    await server_stack.aclose()
                    return name, None

            if transport_type in {"sse", "streamableHttp"}:
                ok, error = validate_url_target(cfg.url)
                if not ok:
                    logger.warning(
                        "MCP server '{}': blocked unsafe URL {} ({})",
                        name,
                        cfg.url,
                        error,
                    )
                    await server_stack.aclose()
                    return name, None

            if transport_type == "stdio":
                command, args, env = _normalize_windows_stdio_command(
                    cfg.command,
                    cfg.args,
                    cfg.env or None,
                )
                params = StdioServerParameters(
                    command=command,
                    args=args,
                    env=env,
                    cwd=cfg.cwd or None,
                )
                read, write = await server_stack.enter_async_context(stdio_client(params))
            elif transport_type == "sse":
                if not await _probe_http_url(cfg.url):
                    logger.warning("MCP server '{}': {} unreachable, skipping", name, cfg.url)
                    await server_stack.aclose()
                    return name, None

                def httpx_client_factory(
                    headers: dict[str, str] | None = None,
                    timeout: httpx.Timeout | None = None,
                    auth: httpx.Auth | None = None,
                ) -> httpx.AsyncClient:
                    merged_headers = {
                        "Accept": "application/json, text/event-stream",
                        **(cfg.headers or {}),
                        **(headers or {}),
                    }
                    return httpx.AsyncClient(
                        headers=merged_headers or None,
                        event_hooks={"request": [_validate_mcp_request_url]},
                        follow_redirects=True,
                        timeout=timeout,
                        auth=auth,
                    )

                read, write = await server_stack.enter_async_context(
                    sse_client(cfg.url, httpx_client_factory=httpx_client_factory)
                )
            elif transport_type == "streamableHttp":
                if not await _probe_http_url(cfg.url):
                    logger.warning("MCP server '{}': {} unreachable, skipping", name, cfg.url)
                    await server_stack.aclose()
                    return name, None

                http_client = await server_stack.enter_async_context(
                    httpx.AsyncClient(
                        headers=cfg.headers or None,
                        event_hooks={"request": [_validate_mcp_request_url]},
                        follow_redirects=True,
                        timeout=None,
                    )
                )
                read, write, _ = await server_stack.enter_async_context(
                    streamable_http_client(cfg.url, http_client=http_client)
                )
            else:
                logger.warning("MCP server '{}': unknown transport type '{}'", name, transport_type)
                await server_stack.aclose()
                return name, None

            read = _filter_malformed_mcp_progress_notifications(read, name)
            session = await server_stack.enter_async_context(ClientSession(read, write))
            await session.initialize()

            tools = await session.list_tools()
            enabled_tools = set(cfg.enabled_tools)
            allow_all_tools = "*" in enabled_tools
            registered_count = 0
            matched_enabled_tools: set[str] = set()
            available_raw_names = [tool_def.name for tool_def in tools.tools]
            available_wrapped_names = [_sanitize_name(f"mcp_{name}_{tool_def.name}") for tool_def in tools.tools]
            for tool_def in tools.tools:
                wrapped_name = _sanitize_name(f"mcp_{name}_{tool_def.name}")
                if (
                    not allow_all_tools
                    and tool_def.name not in enabled_tools
                    and wrapped_name not in enabled_tools
                ):
                    logger.debug(
                        "MCP: skipping tool '{}' from server '{}' (not in enabledTools)",
                        wrapped_name,
                        name,
                    )
                    continue
                wrapper = MCPToolWrapper(session, name, tool_def, tool_timeout=cfg.tool_timeout)
                registry.register(wrapper)
                logger.debug("MCP: registered tool '{}' from server '{}'", wrapper.name, name)
                registered_count += 1
                if enabled_tools:
                    if tool_def.name in enabled_tools:
                        matched_enabled_tools.add(tool_def.name)
                    if wrapped_name in enabled_tools:
                        matched_enabled_tools.add(wrapped_name)

            if enabled_tools and not allow_all_tools:
                unmatched_enabled_tools = sorted(enabled_tools - matched_enabled_tools)
                if unmatched_enabled_tools:
                    logger.warning(
                        "MCP server '{}': enabledTools entries not found: {}. Available raw names: {}. "
                        "Available wrapped names: {}",
                        name,
                        ", ".join(unmatched_enabled_tools),
                        ", ".join(available_raw_names) or "(none)",
                        ", ".join(available_wrapped_names) or "(none)",
                    )

            try:
                resources_result = await session.list_resources()
                for resource in resources_result.resources:
                    wrapper = MCPResourceWrapper(
                        session, name, resource, resource_timeout=cfg.tool_timeout
                    )
                    registry.register(wrapper)
                    registered_count += 1
                    logger.debug(
                        "MCP: registered resource '{}' from server '{}'", wrapper.name, name
                    )
            except Exception as e:
                logger.debug("MCP server '{}': resources not supported or failed: {}", name, e)

            try:
                prompts_result = await session.list_prompts()
                for prompt in prompts_result.prompts:
                    wrapper = MCPPromptWrapper(
                        session, name, prompt, prompt_timeout=cfg.tool_timeout
                    )
                    registry.register(wrapper)
                    registered_count += 1
                    logger.debug("MCP: registered prompt '{}' from server '{}'", wrapper.name, name)
            except Exception as e:
                logger.debug("MCP server '{}': prompts not supported or failed: {}", name, e)

            logger.info(
                "MCP server '{}': connected, {} capabilities registered", name, registered_count
            )
            return name, server_stack

        except Exception as e:
            hint = ""
            text = str(e).lower()
            if any(
                marker in text
                for marker in (
                    "parse error",
                    "invalid json",
                    "unexpected token",
                    "jsonrpc",
                    "content-length",
                )
            ):
                hint = (
                    " Hint: this looks like stdio protocol pollution. Make sure the MCP server writes "
                    "only JSON-RPC to stdout and sends logs/debug output to stderr instead."
                )
            logger.exception("MCP server '{}': failed to connect: {}", name, hint)
            with suppress(Exception):
                await server_stack.aclose()
            return name, None

    server_stacks: dict[str, AsyncExitStack] = {}

    for name, cfg in mcp_servers.items():
        try:
            result = await connect_single_server(name, cfg)
        except Exception as e:
            logger.exception("MCP server '{}' connection failed: {}", name, e)
            continue
        if result is not None and result[1] is not None:
            server_stacks[result[0]] = result[1]

    return server_stacks


def session_extra(metadata: Mapping[str, Any] | None) -> dict[str, Any]:
    """Return persisted session kwargs for MCP preset attachments."""
    mcp_presets = metadata.get("mcp_presets") if isinstance(metadata, Mapping) else None
    return {"mcp_presets": mcp_presets} if isinstance(mcp_presets, list) and mcp_presets else {}


def runtime_lines(
    message: Any,
    *,
    available_server_names: set[str] | None = None,
    configured_server_names: set[str] | None = None,
    connected_server_names: set[str] | None = None,
    skip: bool = False,
) -> list[str]:
    """Return model-visible MCP preset annotations for the current turn."""
    if skip:
        return []
    if configured_server_names is None:
        configured_server_names = available_server_names
    if connected_server_names is None:
        connected_server_names = available_server_names
    metadata = message.metadata if isinstance(getattr(message, "metadata", None), Mapping) else None
    structured = metadata.get("mcp_presets") if isinstance(metadata, Mapping) else None
    if not isinstance(structured, list):
        return []

    lines: list[str] = []
    for item in structured[:8]:
        if not isinstance(item, Mapping):
            continue
        raw_name = str(item.get("name") or "").strip().lower()
        if not raw_name:
            continue
        display = str(item.get("display_name") or raw_name).strip() or raw_name
        transport = str(item.get("transport") or "mcp").strip() or "mcp"
        prefix = f"mcp_{raw_name}_"
        if configured_server_names is not None and raw_name not in configured_server_names:
            lines.append(
                "MCP Preset Attachment: "
                f"@{raw_name} ({display}; transport={transport}) is configured in WebUI Settings, "
                "but this gateway has not loaded the latest MCP settings yet. "
                f"Tools with prefix `{prefix}` may not be available yet; if they are missing, "
                "tell the user to restart nanobot."
            )
            continue
        if connected_server_names is not None and raw_name not in connected_server_names:
            lines.append(
                "MCP Preset Attachment: "
                f"@{raw_name} ({display}; transport={transport}) is configured, "
                "but its MCP connection is not currently live. "
                f"Tools with prefix `{prefix}` may be unavailable; tell the user to open Settings, "
                "run the preset test, and restart nanobot only if hot reload is unavailable."
            )
            continue
        lines.append(
            "MCP Preset Attachment: "
            f"@{raw_name} ({display}; transport={transport}; tool_prefix={prefix}). "
            f"Prefer available tools whose names start with `{prefix}` for this request; "
            "do not substitute shell commands for this MCP integration unless the user asks."
        )
    return lines


async def connect_missing_servers(state: Any, registry: ToolRegistry) -> None:
    """Connect configured MCP servers that are not currently live."""
    missing_servers = {
        name: cfg for name, cfg in state._mcp_servers.items() if name not in state._mcp_stacks
    }
    if state._mcp_connecting or not missing_servers:
        return
    state._mcp_connecting = True
    try:
        connected = await connect_mcp_servers(missing_servers, registry)
        state._mcp_stacks.update(connected)
        _attach_reconnect_handlers(state, registry, connected)
        state._mcp_connected = bool(state._mcp_stacks)
        if connected:
            logger.info("MCP connected servers: {}", sorted(connected))
        else:
            logger.warning("No MCP servers connected successfully (will retry next message)")
    except asyncio.CancelledError:
        logger.warning("MCP connection cancelled (will retry next message)")
        state._mcp_connected = bool(state._mcp_stacks)
    except BaseException as e:
        logger.warning("Failed to connect MCP servers (will retry next message): {}", e)
        state._mcp_connected = bool(state._mcp_stacks)
    finally:
        state._mcp_connecting = False


async def reload_servers(state: Any, registry: ToolRegistry) -> dict[str, Any]:
    """Reconcile live MCP connections with the current config file."""
    async with _reload_lock(state):
        try:
            from nanobot.config.loader import load_config, resolve_config_env_vars

            config = resolve_config_env_vars(load_config())
            next_servers = dict(config.tools.mcp_servers)
        except Exception as exc:
            logger.warning("MCP hot reload could not read config: {}", exc)
            return {
                "ok": False,
                "message": "Could not reload MCP config. Restart nanobot to pick up changes.",
                "requires_restart": True,
                "error": str(exc),
            }

        current_servers = dict(state._mcp_servers)
        current_names = set(current_servers)
        next_names = set(next_servers)
        removed = sorted(current_names - next_names)
        added = sorted(next_names - current_names)
        changed = sorted(
            name
            for name in current_names & next_names
            if _server_signature(current_servers[name]) != _server_signature(next_servers[name])
        )

        tools_removed = 0
        for name in [*removed, *changed]:
            tools_removed += _unregister_server_tools(state, registry, name)
            await _close_server(state, name)

        state._mcp_servers = next_servers
        retry_missing = sorted(
            name
            for name in next_names
            if name not in state._mcp_stacks and name not in set(added) | set(changed)
        )
        to_connect_names = sorted(set(added) | set(changed) | set(retry_missing))
        to_connect = {name: next_servers[name] for name in to_connect_names}
        connected: dict[str, AsyncExitStack] = {}
        if to_connect:
            connected = await connect_mcp_servers(to_connect, registry)
            state._mcp_stacks.update(connected)
            _attach_reconnect_handlers(state, registry, connected)

        state._mcp_connected = bool(state._mcp_stacks)
        failed = sorted(set(to_connect) - set(connected))
        unchanged = not removed and not added and not changed and not retry_missing
        ok = not failed
        if failed:
            message = "MCP config reloaded, but some servers did not connect: " + ", ".join(failed)
        elif unchanged:
            message = "MCP config is already live."
        elif retry_missing and not added and not changed and not removed:
            message = "MCP connections refreshed without restarting nanobot."
        else:
            message = "MCP config reloaded without restarting nanobot."

        logger.info(
            "MCP hot reload: added={} changed={} removed={} retried={} connected={} failed={} tools_removed={}",
            added,
            changed,
            removed,
            retry_missing,
            sorted(connected),
            failed,
            tools_removed,
        )
        return {
            "ok": ok,
            "message": message,
            "added": added,
            "changed": changed,
            "removed": removed,
            "retried": retry_missing,
            "connected": sorted(state._mcp_stacks),
            "configured": sorted(state._mcp_servers),
            "failed": failed,
            "tools_removed": tools_removed,
            "requires_restart": False,
        }


async def request_mcp_reload(bus: Any, *, timeout: float = 15.0) -> dict[str, Any]:
    """Ask the running agent loop to reconcile live MCP connections."""
    loop = asyncio.get_running_loop()
    ack: asyncio.Future[dict[str, Any]] = loop.create_future()
    await bus.publish_inbound(
        InboundMessage(
            channel="system",
            sender_id="webui-settings",
            chat_id="runtime",
            content=RUNTIME_CONTROL_MCP_RELOAD,
            metadata={
                INBOUND_META_RUNTIME_CONTROL: RUNTIME_CONTROL_MCP_RELOAD,
                RUNTIME_CONTROL_ACK: ack,
            },
        )
    )
    try:
        result = await asyncio.wait_for(ack, timeout=timeout)
    except asyncio.TimeoutError:
        return {
            "ok": False,
            "message": "MCP hot reload timed out. Restart nanobot to pick up changes.",
            "requires_restart": True,
        }
    return result if isinstance(result, dict) else {
        "ok": False,
        "message": "MCP hot reload returned an unexpected response.",
        "requires_restart": True,
    }


async def handle_runtime_control(state: Any, msg: InboundMessage, registry: ToolRegistry) -> bool:
    metadata = msg.metadata if isinstance(msg.metadata, dict) else {}
    control = metadata.get(INBOUND_META_RUNTIME_CONTROL)
    if control != RUNTIME_CONTROL_MCP_RELOAD:
        return False

    ack = metadata.get(RUNTIME_CONTROL_ACK)
    try:
        result = await reload_servers(state, registry)
    except Exception as exc:
        logger.exception("MCP hot reload failed")
        result = {
            "ok": False,
            "message": "MCP hot reload failed. Restart nanobot to pick up changes.",
            "requires_restart": True,
            "error": str(exc),
        }
    if isinstance(ack, asyncio.Future) and not ack.done():
        ack.set_result(result)
    return True


def _reload_lock(state: Any) -> asyncio.Lock:
    try:
        return _RELOAD_LOCKS[state]
    except KeyError:
        lock = asyncio.Lock()
        _RELOAD_LOCKS[state] = lock
        return lock


def _attach_reconnect_handlers(
    state: Any,
    registry: ToolRegistry,
    server_names: Mapping[str, Any] | set[str] | list[str] | tuple[str, ...],
) -> None:
    async def reconnect(server_name: str, tool_name: str, stale_tool: Tool) -> Tool | None:
        return await _refresh_terminated_server(
            state,
            registry,
            server_name,
            tool_name,
            stale_tool,
        )

    for server_name in server_names:
        prefix = _tool_prefix(server_name)
        for tool_name in list(registry.tool_names):
            if not tool_name.startswith(prefix):
                continue
            tool = registry.get(tool_name)
            if isinstance(tool, _MCPWrapperBase):
                tool.set_reconnect_handler(reconnect)


async def _refresh_terminated_server(
    state: Any,
    registry: ToolRegistry,
    server_name: str,
    tool_name: str,
    stale_tool: Tool,
) -> Tool | None:
    async with _reload_lock(state):
        cfg = state._mcp_servers.get(server_name)
        if cfg is None:
            logger.warning(
                "MCP server '{}' session terminated but is no longer configured",
                server_name,
            )
            return None

        current_tool = registry.get(tool_name)
        if (
            current_tool is not None
            and current_tool is not stale_tool
            and server_name in state._mcp_stacks
        ):
            return current_tool

        logger.warning("MCP server '{}' session terminated; refreshing connection", server_name)
        _unregister_server_tools(state, registry, server_name)
        await _close_server(state, server_name)

        connected = await connect_mcp_servers({server_name: cfg}, registry)
        state._mcp_stacks.update(connected)
        _attach_reconnect_handlers(state, registry, connected)
        state._mcp_connected = bool(state._mcp_stacks)
        if server_name not in connected:
            logger.warning("MCP server '{}' reconnect failed after session termination", server_name)
            return None
        return registry.get(tool_name)


def _server_signature(cfg: Any) -> Any:
    if hasattr(cfg, "model_dump"):
        return cfg.model_dump(mode="json")
    return cfg


def _tool_prefix(server_name: str) -> str:
    return _sanitize_name(f"mcp_{server_name}_")


def _unregister_server_tools(state: Any, registry: ToolRegistry, server_name: str) -> int:
    prefix = _tool_prefix(server_name)
    removed = 0
    for tool_name in list(registry.tool_names):
        if tool_name.startswith(prefix):
            registry.unregister(tool_name)
            removed += 1
    return removed


async def _close_server(state: Any, server_name: str) -> None:
    stack = state._mcp_stacks.pop(server_name, None)
    if stack is None:
        return
    try:
        await stack.aclose()
    except (RuntimeError, BaseExceptionGroup):
        logger.debug("MCP server '{}' cleanup error (can be ignored)", server_name)