"""MCP client and dynamic tool-provider lifecycle.""" from __future__ import annotations import asyncio import hashlib import json import os import re import shutil import urllib.parse from collections.abc import AsyncIterator, Awaitable, Callable, Iterable, Mapping from contextlib import AsyncExitStack, suppress from typing import TYPE_CHECKING, Any, Literal, Protocol, cast import httpx from loguru import logger from nanobot.agent.tools.base import Tool, ToolResult from nanobot.agent.tools.registry import ToolRegistry from nanobot.security.network import ( PinnedDNSAsyncTransport, env_proxy_applies_to_url, httpx_env_proxy_mounts, resolve_url_target, validate_url_target, ) from nanobot.utils.cancellation import task_is_cancelling if TYPE_CHECKING: from mcp import ClientSession from mcp.types import Prompt, Resource from mcp.types import Tool as MCPToolDefinition from nanobot.agent.tools.mcp_oauth import MCPOAuthHandlers from nanobot.config.schema import Config, MCPServerConfig # 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"_+") _ReconnectCallback = Callable[[str, str, Tool], Awaitable[Tool | None]] MCPServerLoader = Callable[[], Mapping[str, "MCPServerConfig"]] MCPRuntimeStatus = Literal["connecting", "connected", "failed"] class MCPConnection(Protocol): async def aclose(self) -> None: ... async def _close_mcp_connection(name: str, connection: MCPConnection) -> None: try: await connection.aclose() except asyncio.CancelledError: if task_is_cancelling(): raise logger.debug("MCP server '{}' cleanup error (can be ignored)", name) except (RuntimeError, BaseExceptionGroup): logger.debug("MCP server '{}' cleanup error (can be ignored)", name) async def _close_mcp_connections(connections: Mapping[str, MCPConnection]) -> None: cancellation: asyncio.CancelledError | None = None for name, connection in connections.items(): try: await _close_mcp_connection(name, connection) except asyncio.CancelledError as exc: cancellation = cancellation or exc if cancellation is not None: raise cancellation class _OwnedMCPConnection: """Close an MCP transport from the task that originally opened it.""" def __init__(self, owner: asyncio.Task[None], close_requested: asyncio.Event) -> None: self._owner = owner self._close_requested = close_requested async def aclose(self) -> None: self._close_requested.set() try: await asyncio.shield(self._owner) except asyncio.CancelledError: if not self._owner.cancelled(): raise def _is_malformed_mcp_progress_notification(message: Any) -> bool: payload = _mcp_jsonrpc_payload(message) if _payload_value(payload, "method") != "notifications/progress": return False params = _payload_value(payload, "params") return not _progress_params_have_token(params) def _mcp_jsonrpc_payload(message: Any) -> Any: """Return the JSON-RPC payload across current and future MCP SDK shapes.""" envelope = getattr(message, "message", message) return getattr(envelope, "root", None) or envelope def _payload_value(payload: Any, key: str) -> Any: if isinstance(payload, Mapping): return cast(Mapping[str, Any], payload).get(key) return getattr(payload, key, None) def _progress_params_have_token(params: Any) -> bool: if isinstance(params, Mapping): return "progressToken" in params return hasattr(params, "progressToken") or hasattr(params, "progress_token") class _MalformedProgressNotificationFilter: def __init__(self, read_stream: Any, server_name: str) -> None: self._read_stream = read_stream self._server_name = server_name self._iterator: AsyncIterator[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: iterator = self._iterator if iterator is None: iterator = self._read_stream.__aiter__() self._iterator = iterator while True: message = await anext(iterator) 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)) _MAX_TOOL_NAME_LENGTH = 64 _HASH_LENGTH = 8 def _limit_tool_name(name: str, max_length: int = _MAX_TOOL_NAME_LENGTH) -> str: """Limit a tool name while keeping short names unchanged.""" if len(name) <= max_length: return name digest = hashlib.sha1(name.encode("utf-8")).hexdigest()[:_HASH_LENGTH] prefix_length = max_length - _HASH_LENGTH - 1 return f"{name[:prefix_length]}_{digest}" def _sanitize_mcp_tool_name(name: str) -> str: """Sanitize and limit an MCP-derived tool name.""" return _limit_tool_name(_sanitize_name(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_transient_connection_failure(exc: BaseException) -> bool: if isinstance(exc, BaseExceptionGroup): group = cast(BaseExceptionGroup[BaseException], exc) return bool(group.exceptions) and all( _is_transient_connection_failure(nested) for nested in group.exceptions ) return isinstance(exc, (httpx.ConnectError, httpx.ConnectTimeout)) or _is_transient(exc) def _log_mcp_connection_failure(name: str, exc: BaseException, hint: str = "") -> None: if _is_transient_connection_failure(exc): logger.warning("MCP server '{}': transient connection failure", name) logger.opt(exception=exc).debug( "MCP server '{}' transient connection failure details", name ) return logger.opt(exception=exc).error("MCP server '{}': failed to connect: {}", name, hint) def _is_session_terminated(exc: BaseException) -> bool: """Return True when the MCP SDK reports a dead client session.""" if _is_transient(exc): return True 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 ok, _, resolved_ips = resolve_url_target(url) if not ok: return False if env_proxy_applies_to_url(url): return True for target_host in resolved_ips or (host,): try: _reader, writer = await asyncio.wait_for( asyncio.open_connection(target_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): continue return False def _redact_url(url: str) -> str: """Strip credentials and query/fragment before logging an MCP URL. Server URLs may embed secrets (``https://user:token@host/sse`` or a ``?token=`` query). Some deployments also put opaque tokens in the path, so log only the origin and a path placeholder. """ try: parts = urllib.parse.urlsplit(url) hostname = parts.hostname or "" netloc = f"[{hostname}]" if ":" in hostname else hostname if parts.port: netloc = f"{netloc}:{parts.port}" path = "/..." if parts.path and parts.path != "/" else parts.path return urllib.parse.urlunsplit((parts.scheme, netloc, path, "", "")) except Exception: return "" def _pinned_transport_kwargs() -> dict[str, Any]: kwargs: dict[str, Any] = {"transport": PinnedDNSAsyncTransport()} mounts = httpx_env_proxy_mounts() if mounts: kwargs["mounts"] = mounts return kwargs 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 {_redact_url(str(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 cast(list[object], options): if not isinstance(option, dict): return None option_schema = cast(dict[str, Any], option) if option_schema.get("type") == "null": saw_null = True continue non_null.append(option_schema) if saw_null and len(non_null) == 1: return non_null[0], True return None def _resolve_local_schema_ref(root: dict[str, Any], ref: str) -> Any: """Resolve a local JSON Pointer without accepting remote references.""" if not ref.startswith("#"): raise ValueError("not a local JSON Pointer") pointer = urllib.parse.unquote(ref[1:], errors="strict") if not pointer: return root if not pointer.startswith("/"): raise ValueError("not a local JSON Pointer") current: Any = root for raw_part in pointer[1:].split("/"): part = raw_part.replace("~1", "/").replace("~0", "~") if isinstance(current, dict): current = cast(dict[str, Any], current)[part] elif isinstance(current, list): current = cast(list[Any], current)[int(part)] else: raise KeyError(part) return current def _rewrite_local_schema_refs(schema: dict[str, Any]) -> dict[str, Any]: """Hoist arbitrary local JSON-Pointer refs into provider-compatible ``$defs``.""" rewritten_refs: dict[str, str] = {} generated_defs: dict[str, Any] = {} def rewrite(value: Any) -> Any: if isinstance(value, list): return [rewrite(item) for item in cast(list[Any], value)] if not isinstance(value, dict): return value rewritten = dict(cast(dict[str, Any], value)) raw_ref = rewritten.get("$ref") ref = raw_ref if isinstance(raw_ref, str) else None is_rewritable_ref = False if ref is not None and not ref.startswith("#/$defs/"): try: pointer = urllib.parse.unquote(ref[1:], errors="strict") except (UnicodeDecodeError, ValueError): pass else: is_rewritable_ref = ref.startswith("#") and ( not pointer or pointer.startswith("/") ) if is_rewritable_ref: assert ref is not None name = rewritten_refs.get(ref) if name is None: try: target = _resolve_local_schema_ref(schema, ref) except (KeyError, IndexError, TypeError, UnicodeDecodeError, ValueError): logger.warning("MCP tool schema contains an unresolved local $ref: {}", ref) else: name = f"ref_{hashlib.sha256(ref.encode()).hexdigest()[:12]}" existing_defs = schema.get("$defs") while isinstance(existing_defs, dict) and name in existing_defs: name += "_" rewritten_refs[ref] = name # Reserve the name before descending so recursive refs terminate. generated_defs[name] = {} generated_defs[name] = rewrite(target) if name is not None: rewritten["$ref"] = f"#/$defs/{name}" return {key: rewrite(item) for key, item in rewritten.items()} result = cast(dict[str, Any], rewrite(schema)) if generated_defs: existing_defs = result.get("$defs") result["$defs"] = { **(existing_defs if isinstance(existing_defs, dict) else {}), **generated_defs, } return result def _normalize_nullable_schema(schema: dict[str, Any]) -> dict[str, Any]: """Normalize nullable forms in structural subschemas only.""" normalized = dict(schema) raw_type = normalized.get("type") if isinstance(raw_type, list): type_values = cast(list[Any], raw_type) non_null = [item for item in type_values if item != "null"] if "null" in type_values 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 properties = normalized.get("properties") if isinstance(properties, dict): property_schemas = cast(dict[str, Any], properties) normalized["properties"] = { name: ( _normalize_nullable_schema(cast(dict[str, Any], prop)) if isinstance(prop, dict) else prop ) for name, prop in property_schemas.items() } items = normalized.get("items") if isinstance(items, dict): normalized["items"] = _normalize_nullable_schema(cast(dict[str, Any], items)) definitions = normalized.get("$defs") if isinstance(definitions, dict): definition_schemas = cast(dict[str, Any], definitions) normalized["$defs"] = { name: _normalize_nullable_schema(cast(dict[str, Any], definition)) if isinstance(definition, dict) else definition for name, definition in definition_schemas.items() } if normalized.get("type") == "object": normalized.setdefault("properties", {}) normalized.setdefault("required", []) return normalized def _normalize_schema_for_openai(schema: Any) -> dict[str, Any]: """Normalize MCP JSON Schema patterns for tool definitions.""" if not isinstance(schema, dict): return {"type": "object", "properties": {}} schema_mapping = cast(dict[str, Any], schema) return _normalize_nullable_schema(_rewrite_local_schema_refs(schema_mapping)) class _MCPWrapperBase(Tool): """Common reconnect handling for wrappers bound to one MCP server session.""" _plugin_discoverable = False _session: ClientSession _server_name: str _name: str def _set_mcp_connection(self, session: ClientSession, 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 def _image_block_data_url(block: Any, types: Any) -> str | None: """Return a base64 ``data:`` URL for an MCP image-bearing content block. Handles ``ImageContent`` directly and ``EmbeddedResource`` wrapping a binary blob with an ``image/*`` MIME type. Returns ``None`` for anything else. ``getattr`` guards keep this safe when the installed/faked ``mcp`` SDK does not expose a given type. """ image_cls = getattr(types, "ImageContent", None) if image_cls is not None and isinstance(block, image_cls): mime = getattr(block, "mimeType", None) or "image/png" return f"data:{mime};base64,{block.data}" embedded_cls = getattr(types, "EmbeddedResource", None) blob_cls = getattr(types, "BlobResourceContents", None) if embedded_cls is not None and isinstance(block, embedded_cls): resource = getattr(block, "resource", None) if blob_cls is not None and isinstance(resource, blob_cls): blob_resource = cast(Any, resource) mime = getattr(blob_resource, "mimeType", None) or "" if isinstance(mime, str) and mime.startswith("image/"): return f"data:{mime};base64,{blob_resource.blob}" return None def _mcp_image_tool_result(text_parts: list[str], artifacts: list[dict[str, Any]]) -> str: """Build the compact tool result for an MCP call that returned image(s). The base64 stays out of the model context entirely — only artifact paths and metadata are returned, so the result is small and the channel can deliver the saved file via the message tool. """ payload: dict[str, Any] = { "artifacts": artifacts, "next_step": ( "These images were returned by an MCP tool and saved as local artifacts. " "Call the message tool with the artifact 'path' values in the media " "parameter to deliver the images to the user. Do not paste base64 or raw " "paths into your reply unless the user asks for debug details." ), } text = "\n".join(part for part in text_parts if part) if text: payload["text"] = text return json.dumps(payload, ensure_ascii=False) class MCPToolWrapper(_MCPWrapperBase): """Wraps a single MCP server tool as a nanobot Tool.""" _plugin_discoverable = False def __init__( self, session: ClientSession, server_name: str, tool_def: MCPToolDefinition, tool_timeout: int = 30, ): self._set_mcp_connection(session, server_name) self._original_name = tool_def.name self._name = _sanitize_mcp_tool_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: 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 ToolResult.error( 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). if task_is_cancelling(): raise logger.warning("MCP tool '{}' was cancelled by server/SDK", self._name) return ToolResult.error("(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 ToolResult.error( f"(MCP tool call failed after retry: {type(exc).__name__})" ) logger.exception( "MCP tool '{}' failed: {}: {}", self._name, type(exc).__name__, exc, ) return ToolResult.error( f"(MCP tool call failed: {type(exc).__name__})" ) else: # Success — extract text and persist any image content as artifacts. try: rendered = self._render_call_result(result.content, kwargs) if getattr(result, "isError", False): return ToolResult.error(rendered) return rendered except Exception as exc: logger.exception( "MCP tool '{}' failed while rendering result: {}: {}", self._name, type(exc).__name__, exc, ) return ToolResult.error( f"(MCP tool returned malformed content: {type(exc).__name__})" ) def _render_call_result(self, content: Any, arguments: Mapping[str, Any]) -> str: """Turn MCP content blocks into a tool result string. Text is concatenated as before. Image blocks are decoded and saved as local artifacts (mirroring the built-in image generation tool) so the model can deliver them via the message tool instead of trying to forward base64 — which would be truncated and bloat the context window. """ from mcp import types text_parts: list[str] = [] artifacts: list[dict[str, Any]] = [] for block in content: if isinstance(block, types.TextContent): text_parts.append(block.text) continue data_url = _image_block_data_url(block, types) if data_url is not None: stored = self._store_image_block(data_url, arguments) if stored is not None: artifacts.append(stored) else: text_parts.append("(MCP tool returned an image that could not be stored)") continue text_parts.append(str(block)) if artifacts: return _mcp_image_tool_result(text_parts, artifacts) return "\n".join(text_parts) or "(no output)" def _store_image_block( self, data_url: str, arguments: Mapping[str, Any] ) -> dict[str, Any] | None: """Persist one image data URL as an artifact; return its metadata or None.""" from nanobot.utils.artifacts import ArtifactError, store_generated_image_artifact try: return store_generated_image_artifact( data_url, prompt=str(arguments.get("prompt") or ""), model=str(arguments.get("model") or ""), save_dir="generated", provider=f"mcp:{self._server_name}", ) except (ArtifactError, OSError) as exc: logger.warning( "MCP tool '{}' returned an image that could not be stored: {}", self._name, exc, ) return None class MCPResourceWrapper(_MCPWrapperBase): """Wraps an MCP resource URI as a read-only nanobot Tool.""" _plugin_discoverable = False def __init__( self, session: ClientSession, server_name: str, resource_def: Resource, resource_timeout: int = 30, ): self._set_mcp_connection(session, server_name) self._uri = resource_def.uri self._name = _sanitize_mcp_tool_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: if task_is_cancelling(): 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(cast(object, block), types.BlobResourceContents): parts.append(f"[Binary resource: {len(block.blob)} bytes]") else: parts.append(str(block)) return "\n".join(parts) or "(no output)" class MCPPromptWrapper(_MCPWrapperBase): """Wraps an MCP prompt as a read-only nanobot Tool.""" _plugin_discoverable = False def __init__( self, session: ClientSession, server_name: str, prompt_def: Prompt, prompt_timeout: int = 30, ): self._set_mcp_connection(session, server_name) self._prompt_name = prompt_def.name self._name = _sanitize_mcp_tool_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: if task_is_cancelling(): 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)" async def connect_mcp_servers( mcp_servers: dict[str, MCPServerConfig], registry: ToolRegistry, *, oauth_handlers: Mapping[str, MCPOAuthHandlers] | None = None, ) -> dict[str, MCPConnection]: """Connect to configured MCP servers and register their tools, resources, prompts. Returns one connection handle per server. Each handle keeps the task that entered the MCP SDK contexts alive so reconnect and shutdown can close AnyIO cancel scopes from their owning task. """ 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 open_single_server( name: str, cfg: MCPServerConfig, server_stack: AsyncExitStack ) -> bool: 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) return False if transport_type in {"sse", "streamableHttp"}: ok, error = validate_url_target(cfg.url) if not ok: logger.warning( "MCP server '{}': blocked unsafe URL {} ({})", name, _redact_url(cfg.url), error, ) return False oauth_auth: httpx.Auth | None = None if cfg.auth == "oauth": if transport_type not in {"sse", "streamableHttp"}: logger.warning( "MCP server '{}': OAuth requires an SSE or Streamable HTTP transport", name, ) return False from nanobot.agent.tools.mcp_oauth import ( MCPAuthorizationRequiredError, create_mcp_oauth_auth, ) try: oauth_auth = await create_mcp_oauth_auth( name, cfg.url, (oauth_handlers or {}).get(name), ) except MCPAuthorizationRequiredError: logger.info("MCP server '{}': waiting for browser authorization", name) return False 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, _redact_url(cfg.url)) return False 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, **_pinned_transport_kwargs(), ) sse_kwargs: dict[str, Any] = { "httpx_client_factory": httpx_client_factory, } if oauth_auth is not None: sse_kwargs["auth"] = oauth_auth read, write = await server_stack.enter_async_context( sse_client(cfg.url, **sse_kwargs) ) elif transport_type == "streamableHttp": if not await _probe_http_url(cfg.url): logger.warning("MCP server '{}': {} unreachable, skipping", name, _redact_url(cfg.url)) return False http_client_kwargs: dict[str, Any] = { "headers": cfg.headers or None, "event_hooks": {"request": [_validate_mcp_request_url]}, "follow_redirects": True, "timeout": httpx.Timeout(30.0, connect=10.0), **_pinned_transport_kwargs(), } if oauth_auth is not None: http_client_kwargs["auth"] = oauth_auth http_client = await server_stack.enter_async_context( httpx.AsyncClient(**http_client_kwargs) ) 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) return False 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_mcp_tool_name(f"mcp_{name}_{tool_def.name}") for tool_def in tools.tools] for tool_def in tools.tools: wrapped_name = _sanitize_mcp_tool_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)", ) # Only register resources and prompts when no tool restriction is # active. enabledTools is a per-*tool* allowlist; resources and # prompts have no equivalent name filter, so they must be skipped # whenever the operator specified a tool subset. An empty list # (deny-all) or a list of specific tool names both indicate that # the operator intended to restrict capabilities — registering # unrestricted resource/prompt wrappers would violate that intent. # The default ["*"] (allow-all) means no restriction was intended. register_extras = allow_all_tools if register_extras: 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 ) else: logger.info( "MCP server '{}': skipping resource/prompt registration " "(enabledTools does not include '*' — only tools allowed)", name, ) logger.info( "MCP server '{}': connected, {} capabilities registered", name, registered_count ) return True 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." ) _log_mcp_connection_failure(name, e, hint) return False async def connect_single_server( name: str, cfg: MCPServerConfig ) -> tuple[str, MCPConnection | None]: loop = asyncio.get_running_loop() ready: asyncio.Future[bool] = loop.create_future() close_requested = asyncio.Event() async def own_connection() -> None: try: async with AsyncExitStack() as stack: connected = await open_single_server(name, cfg, stack) if not ready.done(): ready.set_result(connected) if connected: await close_requested.wait() except BaseException as exc: if not ready.done(): ready.set_exception(exc) raise owner = asyncio.create_task(own_connection(), name=f"mcp:{name}") connection = _OwnedMCPConnection(owner, close_requested) try: connected = await ready except BaseException as exc: close_requested.set() owner.cancel() with suppress(BaseException): await asyncio.shield(owner) if isinstance(exc, asyncio.CancelledError) and not task_is_cancelling(): logger.warning("MCP server '{}': connection cancelled by server/SDK", name) return name, None raise if not connected: await connection.aclose() return name, None return name, connection server_stacks: dict[str, MCPConnection] = {} attempted_names: list[str] = [] try: for name, cfg in mcp_servers.items(): attempted_names.append(name) try: result = await connect_single_server(name, cfg) except Exception as e: _log_mcp_connection_failure(name, e) continue if result[1] is not None: server_stacks[result[0]] = result[1] except BaseException: # Callers can bound readiness/reload with a timeout. If cancellation # interrupts a later server, ownership of earlier connections has not # transferred yet, so roll the whole batch back before propagating it. for name in attempted_names: _unregister_server_tools(registry, name) try: await _close_mcp_connections(server_stacks) except BaseException as cleanup_exc: logger.debug("MCP batch rollback cleanup error (can be ignored): {}", cleanup_exc) raise 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 _configured_servers(config: Config) -> dict[str, MCPServerConfig]: from nanobot.agent.plugins import agent_plugin_mcp_servers return agent_plugin_mcp_servers( config.workspace_path, config.tools.mcp_servers, ) def _load_current_servers() -> dict[str, MCPServerConfig]: from nanobot.config.loader import load_config, resolve_config_env_vars return _configured_servers(resolve_config_env_vars(load_config())) class MCPProvider: """Own configured MCP connections and their dynamic tool registrations.""" def __init__( self, servers: Mapping[str, MCPServerConfig], registry: ToolRegistry, *, server_loader: MCPServerLoader | None = None, ) -> None: self._servers = dict(servers) self._registry = registry self._server_loader = server_loader or _load_current_servers self._connections: dict[str, MCPConnection] = {} self._runtime_statuses: dict[str, MCPRuntimeStatus] = {} self._lock = asyncio.Lock() self._closing = False @classmethod def from_config( cls, config: Config, registry: ToolRegistry, *, server_loader: MCPServerLoader | None = None, ) -> MCPProvider: return cls( _configured_servers(config), registry, server_loader=server_loader, ) @property def configured_server_names(self) -> set[str]: return set(self._servers) @property def connected_server_names(self) -> set[str]: return set(self._connections) def runtime_status(self) -> dict[str, MCPRuntimeStatus]: """Return the latest connection-attempt result for configured servers.""" return { name: status for name, status in self._runtime_statuses.items() if name in self._servers } def _set_runtime_status( self, server_names: Iterable[str], status: MCPRuntimeStatus, ) -> None: for name in server_names: self._runtime_statuses[name] = status def _record_connection_result( self, attempted: Iterable[str], connected: Iterable[str], ) -> None: attempted_names = set(attempted) connected_names = set(connected) self._set_runtime_status(connected_names, "connected") self._set_runtime_status(attempted_names - connected_names, "failed") async def connect(self) -> None: """Connect configured servers that are not currently live.""" async with self._lock: if self._closing: return configured_missing = { name: cfg for name, cfg in self._servers.items() if name not in self._connections } oauth_servers = { name: cfg for name, cfg in configured_missing.items() if cfg.auth == "oauth" } authorization_pending: set[str] = set() if oauth_servers: from nanobot.agent.tools.mcp_oauth import mcp_oauth_has_credentials authorization_pending = { name for name, cfg in oauth_servers.items() if not mcp_oauth_has_credentials(name, cfg.url) } for name in authorization_pending: self._runtime_statuses.pop(name, None) missing_servers = { name: cfg for name, cfg in configured_missing.items() if name not in authorization_pending } if not missing_servers: return self._set_runtime_status(missing_servers, "connecting") try: connected = await connect_mcp_servers(missing_servers, self._registry) if self._closing: await _close_mcp_connections(connected) return self._connections.update(connected) self._record_connection_result(missing_servers, connected) self._attach_reconnect_handlers(connected) if connected: logger.info("MCP connected servers: {}", sorted(connected)) else: logger.warning( "No MCP servers connected successfully " "(will retry on the next readiness check)" ) except asyncio.CancelledError: self._set_runtime_status(missing_servers, "failed") if task_is_cancelling(): raise logger.warning( "MCP connection cancelled (will retry on the next readiness check)" ) except BaseException as exc: self._set_runtime_status(missing_servers, "failed") logger.warning( "Failed to connect MCP servers " "(will retry on the next readiness check): {}", exc, ) async def reload(self) -> dict[str, Any]: """Reconcile live MCP connections with the current configuration.""" async with self._lock: if self._closing: return self._closing_result() try: next_servers = dict(self._server_loader()) 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(self._servers) current_names = set(current_servers) next_names = set(next_servers) from nanobot.agent.tools.mcp_oauth import mcp_oauth_has_credentials authorization_pending = { name for name, cfg in next_servers.items() if cfg.auth == "oauth" and not mcp_oauth_has_credentials(name, cfg.url) } 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(self._registry, name) await self._close_server(name) for name in [*removed, *authorization_pending]: self._runtime_statuses.pop(name, None) self._servers = next_servers retry_missing = sorted( name for name in next_names if name not in self._connections and name not in set(added) | set(changed) and name not in authorization_pending ) to_connect_names = sorted( (set(added) | set(changed) | set(retry_missing)) - authorization_pending ) to_connect = {name: next_servers[name] for name in to_connect_names} connected: dict[str, MCPConnection] = {} if to_connect: self._set_runtime_status(to_connect, "connecting") try: connected = await connect_mcp_servers(to_connect, self._registry) except BaseException: self._set_runtime_status(to_connect, "failed") raise if self._closing: await _close_mcp_connections(connected) return self._closing_result() self._connections.update(connected) self._record_connection_result(to_connect, connected) self._attach_reconnect_handlers(connected) 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(self._connections), "configured": sorted(self._servers), "failed": failed, "tools_removed": tools_removed, "requires_restart": False, } @staticmethod def _closing_result() -> dict[str, Any]: return { "ok": False, "message": "MCP connections are shutting down.", "requires_restart": True, } def _attach_reconnect_handlers(self, server_names: Iterable[str]) -> None: async def reconnect( server_name: str, tool_name: str, stale_tool: Tool, ) -> Tool | None: return await self._refresh_terminated_server( server_name, tool_name, stale_tool, ) for server_name in server_names: for tool_name in list(self._registry.tool_names): tool = self._registry.get(tool_name) if not _tool_belongs_to_server(tool, tool_name, server_name): continue if isinstance(tool, _MCPWrapperBase): tool.set_reconnect_handler(reconnect) async def _refresh_terminated_server( self, server_name: str, tool_name: str, stale_tool: Tool, ) -> Tool | None: async with self._lock: if self._closing: return None cfg = self._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 = self._registry.get(tool_name) if ( current_tool is not None and current_tool is not stale_tool and server_name in self._connections ): return current_tool logger.warning( "MCP server '{}' session terminated; refreshing connection", server_name, ) _unregister_server_tools(self._registry, server_name) await self._close_server(server_name) self._set_runtime_status({server_name}, "connecting") connected = await connect_mcp_servers( {server_name: cfg}, self._registry, ) if self._closing: await _close_mcp_connections(connected) return None self._connections.update(connected) self._record_connection_result({server_name}, connected) self._attach_reconnect_handlers(connected) if server_name not in connected: logger.warning( "MCP server '{}' reconnect failed after session termination", server_name, ) return None return self._registry.get(tool_name) async def _close_server(self, server_name: str) -> None: connection = self._connections.pop(server_name, None) if connection is None: return await _close_mcp_connection(server_name, connection) async def aclose(self) -> None: """Close every connection while excluding reconnect and hot reload.""" self._closing = True async with self._lock: connections = dict(self._connections) self._connections.clear() self._runtime_statuses.clear() for name in self._servers: _unregister_server_tools(self._registry, name) await _close_mcp_connections(connections) 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 _tool_belongs_to_server(tool: Tool | None, tool_name: str, server_name: str) -> bool: if isinstance(tool, _MCPWrapperBase): return getattr(tool, "_server_name", None) == server_name return tool_name.startswith(_tool_prefix(server_name)) def _unregister_server_tools(registry: ToolRegistry, server_name: str) -> int: removed = 0 for tool_name in list(registry.tool_names): tool = registry.get(tool_name) if _tool_belongs_to_server(tool, tool_name, server_name): registry.unregister(tool_name) removed += 1 return removed