Skip to main content
Glama
Rootly-AI-Labs

Rootly MCP server

Official

createService

Create and configure new services with attributes like name, description, and associated IDs for integrations such as PagerDuty, Opsgenie, and GitHub.

Instructions

Creates a new service from provided data

Responses:

  • 201 (Success): service created

    • Content-Type: application/vnd.api+json

    • Example:

{
  "key": "value"
}
  • 401: responds with unauthorized for invalid token

    • Content-Type: application/vnd.api+json

    • Example:

{
  "key": "value"
}
  • 422: invalid request

    • Content-Type: application/vnd.api+json

    • Example:

{
  "key": "value"
}

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
dataYes

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault

No arguments

Implementation Reference

  • Registers the 'createService' tool (and all other OpenAPI-derived tools) using FastMCP.from_openapi with the filtered Rootly API spec containing the /services POST endpoint.
    # Create the MCP server using OpenAPI integration
    # By default, all routes become tools which is what we want
    mcp = FastMCP.from_openapi(
        openapi_spec=filtered_spec,
        client=http_client.client,
        name=name,
        timeout=30.0,
        tags={"rootly", "incident-management"},
    )
  • Defines the allowed API paths including '/services' (for createService POST) and '/services/{service_id}' which are filtered into the OpenAPI spec for tool generation.
    DEFAULT_ALLOWED_PATHS = [
        "/incidents/{incident_id}/alerts",
        "/alerts",
        "/alerts/{alert_id}",
        "/severities",
        "/severities/{severity_id}",
        "/teams",
        "/teams/{team_id}",
        "/services",
        "/services/{service_id}",
        "/functionalities",
        "/functionalities/{functionality_id}",
        # Incident types
        "/incident_types",
        "/incident_types/{incident_type_id}",
        # Action items (all, by id, by incident)
        "/incident_action_items",
        "/incident_action_items/{incident_action_item_id}",
        "/incidents/{incident_id}/action_items",
        # Workflows
        "/workflows",
        "/workflows/{workflow_id}",
        # Workflow runs
        "/workflow_runs",
        "/workflow_runs/{workflow_run_id}",
        # Environments
        "/environments",
        "/environments/{environment_id}",
        # Users
        "/users",
        "/users/{user_id}",
        "/users/me",
        # Status pages
        "/status_pages",
        "/status_pages/{status_page_id}",
        # On-call schedules and shifts
        "/schedules",
        "/schedules/{schedule_id}",
        "/schedules/{schedule_id}/shifts",
        "/shifts",
        "/schedule_rotations/{schedule_rotation_id}",
        "/schedule_rotations/{schedule_rotation_id}/schedule_rotation_users",
        "/schedule_rotations/{schedule_rotation_id}/schedule_rotation_active_days",
        # On-call overrides
        "/schedules/{schedule_id}/override_shifts",
        "/override_shifts/{override_shift_id}",
        # On-call shadows and roles
        "/schedules/{schedule_id}/on_call_shadows",
        "/on_call_shadows/{on_call_shadow_id}",
        "/on_call_roles",
        "/on_call_roles/{on_call_role_id}",
    ]
  • Custom HTTP client used by all generated tools (including createService) to make authenticated requests to the Rootly API, with parameter transformation.
    class AuthenticatedHTTPXClient:
        """An HTTPX client wrapper that handles Rootly API authentication and parameter transformation."""
    
        def __init__(self, base_url: str = "https://api.rootly.com", hosted: bool = False, parameter_mapping: Optional[Dict[str, str]] = None):
            self._base_url = base_url
            self.hosted = hosted
            self._api_token = None
            self.parameter_mapping = parameter_mapping or {}
    
            if not self.hosted:
                self._api_token = self._get_api_token()
    
            # Create the HTTPX client  
            headers = {
                "Content-Type": "application/vnd.api+json", 
                "Accept": "application/vnd.api+json"
                # Let httpx handle Accept-Encoding automatically with all supported formats
            }
            if self._api_token:
                headers["Authorization"] = f"Bearer {self._api_token}"
    
            self.client = httpx.AsyncClient(
                base_url=base_url,
                headers=headers,
                timeout=30.0,
                follow_redirects=True,
                # Ensure proper handling of compressed responses
                limits=httpx.Limits(max_keepalive_connections=5, max_connections=10)
            )
    
        def _get_api_token(self) -> Optional[str]:
            """Get the API token from environment variables."""
            api_token = os.getenv("ROOTLY_API_TOKEN")
            if not api_token:
                logger.warning("ROOTLY_API_TOKEN environment variable is not set")
                return None
            return api_token
    
        def _transform_params(self, params: Optional[Dict[str, Any]]) -> Optional[Dict[str, Any]]:
            """Transform sanitized parameter names back to original names."""
            if not params or not self.parameter_mapping:
                return params
    
            transformed = {}
            for key, value in params.items():
                # Use the original name if we have a mapping, otherwise keep the sanitized name
                original_key = self.parameter_mapping.get(key, key)
                transformed[original_key] = value
                if original_key != key:
                    logger.debug(f"Transformed parameter: '{key}' -> '{original_key}'")
            return transformed
    
        async def request(self, method: str, url: str, **kwargs):
            """Override request to transform parameters."""
            # Transform query parameters
            if 'params' in kwargs:
                kwargs['params'] = self._transform_params(kwargs['params'])
    
            # Call the underlying client's request method and let it handle everything
            return await self.client.request(method, url, **kwargs)
    
        async def get(self, url: str, **kwargs):
            """Proxy to request with GET method."""
            return await self.request('GET', url, **kwargs)
    
        async def post(self, url: str, **kwargs):
            """Proxy to request with POST method."""
            return await self.request('POST', url, **kwargs)
    
        async def put(self, url: str, **kwargs):
            """Proxy to request with PUT method."""
            return await self.request('PUT', url, **kwargs)
    
        async def patch(self, url: str, **kwargs):
            """Proxy to request with PATCH method."""
            return await self.request('PATCH', url, **kwargs)
    
        async def delete(self, url: str, **kwargs):
            """Proxy to request with DELETE method."""
            return await self.request('DELETE', url, **kwargs)
    
        async def __aenter__(self):
            return self
    
        async def __aexit__(self, exc_type, exc_val, exc_tb):
            pass
    
        def __getattr__(self, name):
            # Delegate all other attributes to the underlying client, except for request methods
            if name in ['request', 'get', 'post', 'put', 'patch', 'delete']:
                # Use our overridden methods instead
                return getattr(self, name)
            return getattr(self.client, name)
        
        @property 
        def base_url(self):
            return self._base_url
            
        @property
        def headers(self):
            return self.client.headers
  • Sanitizes parameter names in the OpenAPI spec to MCP-compliant format before tool generation; used for createService schema.
    def sanitize_parameters_in_spec(spec: Dict[str, Any]) -> Dict[str, str]:
        """
        Sanitize all parameter names in an OpenAPI specification.
        
        This function modifies the spec in-place and builds a mapping 
        of sanitized names to original names.
        
        Args:
            spec: OpenAPI specification dictionary
            
        Returns:
            Dictionary mapping sanitized names to original names
        """
        parameter_mapping = {}
        
        # Sanitize parameters in paths
        if "paths" in spec:
            for path, path_item in spec["paths"].items():
                if not isinstance(path_item, dict):
                    continue
                    
                # Sanitize path-level parameters
                if "parameters" in path_item:
                    for param in path_item["parameters"]:
                        if "name" in param:
                            original_name = param["name"]
                            sanitized_name = sanitize_parameter_name(original_name)
                            if sanitized_name != original_name:
                                logger.debug(f"Sanitized path-level parameter: '{original_name}' -> '{sanitized_name}'")
                                param["name"] = sanitized_name
                                parameter_mapping[sanitized_name] = original_name
                
                # Sanitize operation-level parameters
                for method, operation in path_item.items():
                    if method.lower() not in ["get", "post", "put", "delete", "patch", "options", "head", "trace"]:
                        continue
                    if not isinstance(operation, dict):
                        continue
                        
                    if "parameters" in operation:
                        for param in operation["parameters"]:
                            if "name" in param:
                                original_name = param["name"]
                                sanitized_name = sanitize_parameter_name(original_name)
                                if sanitized_name != original_name:
                                    logger.debug(f"Sanitized operation parameter: '{original_name}' -> '{sanitized_name}'")
                                    param["name"] = sanitized_name
                                    parameter_mapping[sanitized_name] = original_name
        
        # Sanitize parameters in components (OpenAPI 3.0)
        if "components" in spec and "parameters" in spec["components"]:
            for param_name, param_def in spec["components"]["parameters"].items():
                if isinstance(param_def, dict) and "name" in param_def:
                    original_name = param_def["name"]
                    sanitized_name = sanitize_parameter_name(original_name)
                    if sanitized_name != original_name:
                        logger.debug(f"Sanitized component parameter: '{original_name}' -> '{sanitized_name}'")
                        param_def["name"] = sanitized_name
                        parameter_mapping[sanitized_name] = original_name
        
        return parameter_mapping
  • Internal async request function used by custom tools, but generated tools use the http_client directly via FastMCP.
    """Make an authenticated request, extracting token from MCP headers in hosted mode."""
    # In hosted mode, get token from MCP request headers
    if hosted:
        try:
            from fastmcp.server.dependencies import get_http_headers
            request_headers = get_http_headers()
            auth_header = request_headers.get("authorization", "")
            if auth_header:
                # Add authorization header to the request
                if "headers" not in kwargs:
                    kwargs["headers"] = {}
                kwargs["headers"]["Authorization"] = auth_header
        except Exception:
            pass  # Fallback to default client behavior
    
    # Use our custom client with proper error handling instead of bypassing it
    return await http_client.request(method, url, **kwargs)
Behavior2/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

With no annotations provided, the description carries the full burden of behavioral disclosure. It mentions HTTP response codes (201, 401, 422) and content types, which adds some context about error handling and success conditions. However, it doesn't describe critical behavioral aspects like authentication requirements, rate limits, side effects, or what 'service' means in this context, leaving significant gaps for a mutation tool.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is moderately structured with a brief purpose statement followed by HTTP response details. However, the response examples are generic placeholders ('key': 'value') that don't add value, and the formatting with markdown and code blocks makes it less concise than it could be. The front-loaded purpose statement is good, but the response details could be more efficiently presented.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness3/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity (1 parameter with nested objects, no annotations, but has output schema), the description is partially complete. It covers basic purpose and response codes, but lacks crucial context about the 'service' domain, input requirements, and behavioral traits. The existence of an output schema reduces the need to describe return values, but the description still misses important operational context for a creation tool.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters2/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The schema description coverage is 0%, meaning none of the parameters have descriptions in the schema itself. The tool description mentions 'provided data' but provides no details about what data is needed, what a 'service' consists of, or how to structure the input. This fails to compensate for the complete lack of schema documentation, leaving parameters largely unexplained.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose4/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose as 'Creates a new service from provided data', which is a specific verb+resource combination. However, it doesn't differentiate from sibling tools like 'createEnvironment', 'createTeam', or 'createWorkflow', which all follow the same 'create X from provided data' pattern, so it doesn't fully distinguish itself from alternatives.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines2/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

The description provides no guidance on when to use this tool versus alternatives like 'listServices' or other creation tools. It doesn't mention prerequisites, dependencies, or contextual factors that would help an agent decide between this and sibling tools like 'createEnvironment' or 'createTeam'.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Install Server

Other Tools

Related Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/Rootly-AI-Labs/Rootly-MCP-server'

If you have feedback or need assistance with the MCP directory API, please join our Discord server