Una herramienta basada en FastMCP para interactuar con Splunk Enterprise/Cloud mediante lenguaje natural. Esta herramienta ofrece un conjunto de funciones para buscar datos de Splunk, gestionar almacenes KV y acceder a recursos de Splunk mediante una interfaz intuitiva.
La herramienta funciona en tres modos:
Modo SSE (predeterminado)
Comunicación basada en eventos enviados por el servidor
Interacción bidireccional en tiempo real
Adecuado para clientes MCP basados en web
Modo predeterminado cuando no se proporcionan argumentos
Acceso a través del punto final /sse
Modo API
Puntos finales de API RESTful
Acceso a través del prefijo de punto final /api/v1
Comience con python splunk_mcp.py api
Modo STDIO
Comunicación basada en entrada/salida estándar
Compatible con Claude Desktop y otros clientes MCP
Ideal para la integración directa con asistentes de IA
Comience con python splunk_mcp.py stdio
Related MCP server: MCP Server for Splunk
Búsqueda de Splunk : ejecute búsquedas de Splunk con consultas en lenguaje natural
Gestión de índices : enumerar e inspeccionar índices de Splunk
Administración de usuarios : ver y administrar usuarios de Splunk
Operaciones de la tienda KV : crear, enumerar y administrar colecciones de la tienda KV
Compatibilidad asincrónica : creada con patrones asíncronos/en espera para un mejor rendimiento
Registro detallado : registro completo con indicadores emoji para una mejor visibilidad
Configuración SSL : opciones de verificación SSL flexibles para diferentes requisitos de seguridad
Depuración mejorada : conexión detallada y registro de errores para la resolución de problemas
Pruebas integrales : pruebas unitarias que cubren todas las funcionalidades principales
Manejo de errores : Manejo de errores robusto con códigos de estado apropiados
Cumplimiento de SSE : Totalmente compatible con la especificación SSE de MCP
Las siguientes herramientas están disponibles a través de la interfaz MCP:
lista_de_herramientas
Enumera todas las herramientas MCP disponibles con sus descripciones y parámetros
chequeo de salud
Devuelve una lista de aplicaciones Splunk disponibles para verificar la conectividad
silbido
Punto final de ping simple para verificar que el servidor MCP esté activo
usuario_actual
Devuelve información sobre el usuario actualmente autenticado
lista_usuarios
Devuelve una lista de todos los usuarios y sus roles.
índices de lista
Devuelve una lista de todos los índices de Splunk accesibles
obtener_información_de_índice
Devuelve información detallada sobre un índice específico
Parámetros: index_name (cadena)
índices_y_tipos_de_fuente
Devuelve una lista completa de índices y sus tipos de origen.
búsqueda_splunk
Ejecuta una consulta de búsqueda de Splunk
Parámetros:
search_query (cadena): cadena de búsqueda de Splunk
earliest_time (cadena, opcional): hora de inicio de la ventana de búsqueda
latest_time (cadena, opcional): hora de finalización de la ventana de búsqueda
max_results (entero, opcional): número máximo de resultados a devolver
lista_de_búsquedas_guardadas
Devuelve una lista de búsquedas guardadas en la instancia de Splunk
lista_colecciones_kvstore
Enumera todas las colecciones de la tienda KV
crear_colección_kvstore
Crea una nueva colección de tienda KV
Parámetros: collection_name (cadena)
eliminar_colección_kvstore
Elimina una colección de tienda KV existente
Parámetros: collection_name (cadena)
Al ejecutarse en modo SSE, están disponibles los siguientes puntos finales:
/sse : Devuelve información de conexión SSE en formato de texto/flujo de eventos
Proporciona metadatos sobre la conexión SSE
Incluye URL para el punto final de los mensajes
Proporciona información sobre protocolos y capacidades.
/sse/messages : El punto final principal del flujo SSE
Transmite eventos del sistema como latidos
Mantiene una conexión persistente
Envía eventos SSE con el formato correcto
/sse/health : Punto final de verificación de estado para el modo SSE
Devuelve información de estado y versión en formato SSE
La implementación de MCP incluye un manejo de errores consistente:
Comandos de búsqueda no válidos o solicitudes mal formadas
Permisos insuficientes
Recurso no encontrado
Validación de entrada no válida
Errores inesperados del servidor
Problemas de conexión con el servidor Splunk
Todas las respuestas de error incluyen un mensaje detallado que explica el error.
Python 3.10 o superior
Poesía para la gestión de la dependencia
Instancia de Splunk Enterprise/Cloud
Credenciales de Splunk adecuadas con los permisos necesarios
Clonar el repositorio:
git clone <repository-url>
cd splunk-mcpInstalar dependencias usando Poetry:
poetry installCopie el archivo de entorno de ejemplo y configure sus ajustes:
cp .env.example .envActualice el archivo .env con sus credenciales de Splunk:
SPLUNK_HOST=your_splunk_host
SPLUNK_PORT=8089
SPLUNK_USERNAME=your_username
SPLUNK_PASSWORD=your_password
SPLUNK_SCHEME=https
VERIFY_SSL=true
FASTMCP_LOG_LEVEL=INFOExtrae la última imagen:
docker pull livehybrid/splunk-mcp:latestCree su archivo .env como se indica arriba o utilice variables de entorno directamente.
Ejecutar usando Docker Compose:
docker-compose up -dO usando Docker directamente:
docker run -i \
--env-file .env \
livehybrid/splunk-mcpLa herramienta puede funcionar en tres modos:
Modo SSE (predeterminado para clientes MCP):
# Start in SSE mode (default)
poetry run python splunk_mcp.py
# or explicitly:
poetry run python splunk_mcp.py sse
# Use uvicorn directly:
SERVER_MODE=api poetry run uvicorn splunk_mcp:app --host 0.0.0.0 --port 8000 --reloadModo STDIO:
poetry run python splunk_mcp.py stdioEl proyecto admite tanto el nuevo comando docker compose (V2) como el antiguo docker-compose (V1). Los ejemplos a continuación utilizan la sintaxis V2, pero ambos son compatibles.
Modo SSE (predeterminado):
docker compose up -d mcpModo API:
docker compose run --rm mcp python splunk_mcp.py apiModo STDIO:
docker compose run -i --rm mcp python splunk_mcp.py stdioEl proyecto incluye un entorno de prueba dedicado en Docker:
Ejecutar todas las pruebas:
./run_tests.sh --dockerEjecutar componentes de prueba específicos:
# Run only the MCP server
docker compose up -d mcp
# Run only the test container
docker compose up test
# Run both with test results
docker compose up --abort-on-container-exitLos resultados de la prueba estarán disponibles en el directorio ./test-results .
Imágenes del edificio :
# Build both images
docker compose build
# Build specific service
docker compose build mcp
docker compose build testVisualización de registros :
# View all logs
docker compose logs
# Follow specific service logs
docker compose logs -f mcpDepuración :
# Run with debug mode
DEBUG=true docker compose up mcp
# Access container shell
docker compose exec mcp /bin/bashNota: Si está utilizando Docker Compose V1, reemplace docker compose con docker-compose en los comandos anteriores.
Variables de entorno :
Nunca confirmes archivos .env
Utilice .env.example como plantilla
Considere usar secretos de Docker para producción
Verificación SSL :
VERIFY_SSL=true recomendado para producción
Se puede deshabilitar para desarrollo/pruebas
Configurar a través de variables de entorno
Exposición del puerto :
Exponer únicamente los puertos necesarios
Utilice la red interna de Docker cuando sea posible
Considere la seguridad de la red en la producción
Configure las siguientes variables de entorno:
SPLUNK_HOST : Su dirección de host de Splunk
SPLUNK_PORT : puerto de administración de Splunk (predeterminado: 8089)
SPLUNK_USERNAME : Su nombre de usuario de Splunk
SPLUNK_PASSWORD : Su contraseña de Splunk
SPLUNK_SCHEME : Esquema de conexión (predeterminado: https)
VERIFY_SSL : Habilitar o deshabilitar la verificación SSL (valor predeterminado: verdadero)
FASTMCP_LOG_LEVEL : Nivel de registro (predeterminado: INFO)
SERVER_MODE : Modo servidor (sse, api, stdio) al usar uvicorn
La herramienta proporciona opciones flexibles de verificación SSL:
Modo predeterminado (seguro) :
VERIFY_SSL=trueVerificación completa del certificado SSL
Verificación del nombre de host habilitada
Recomendado para entornos de producción
Modo relajado :
VERIFY_SSL=falseVerificación del certificado SSL deshabilitada
Verificación del nombre de host deshabilitada
Útil para probar certificados autofirmados.
El proyecto incluye una cobertura de pruebas integral utilizando pytest y pruebas de extremo a extremo con un cliente MCP personalizado:
Ejecución de prueba básica:
poetry run pytestCon informes de cobertura:
poetry run pytest --cov=splunk_mcpCon salida detallada:
poetry run pytest -vEl proyecto incluye un script de prueba de cliente MCP personalizado que se conecta al punto final SSE en vivo y prueba todas las herramientas:
# Test all tools
python test_endpoints.py
# Test specific tools
python test_endpoints.py health_check list_indexes
# List all available tools
python test_endpoints.py --listEste script actúa como un cliente MCP mediante:
Conectarse al punto final /sse para obtener la URL de los mensajes
Envío de invocaciones de herramientas al punto final de mensajes
Procesamiento de los eventos SSE para extraer resultados de la herramienta
Validación de los resultados frente a los formatos esperados
Esto proporciona una prueba real de la interfaz SSE tal como la utilizaría un cliente MCP real.
El proyecto utiliza tres enfoques de prueba complementarios:
Pruebas de integración de MCP ( tests/test_api.py ) :
Prueba la interfaz de herramientas MCP a través de mcp.call_tool()
Verifica el registro correcto de la herramienta con FastMCP
Garantiza el formato de respuesta y la estructura de datos correctos.
Valida el manejo de errores en el nivel de interfaz MCP
Nota: Lo ideal sería cambiar el nombre de este archivo a test_mcp.py para reflejar mejor su propósito.
Pruebas de función directa ( tests/test_endpoints_pytest.py ) :
Prueba las funciones de Splunk directamente (omitiendo la capa MCP)
Proporciona una cobertura más completa de los detalles de implementación de funciones.
Prueba casos extremos, variaciones de parámetros y manejo de errores.
Incluye pruebas de configuración de SSL, parámetros de conexión y tiempos de espera.
Utiliza pruebas parametrizadas para una cobertura de pruebas eficiente
Pruebas de cliente MCP de extremo a extremo ( test_endpoints.py ) :
Se comporta como un cliente MCP real que se conecta al punto final SSE
Prueba el flujo completo desde la conexión hasta la invocación de la herramienta y el análisis de la respuesta.
Valida la implementación real del protocolo SSE
Prueba herramientas con parámetros reales contra el servidor en vivo
Pruebas de configuración ( tests/test_config.py ) :
Pruebas para el análisis de variables de entorno
Configuración de verificación SSL
Validación de parámetros de conexión
Las pruebas admiten:
Pruebas asincrónicas con pytest-asyncio
Informes de cobertura con pytest-cov
Burlándose con pytest-mock
Pruebas parametrizadas
Prueba de tiempo de espera de conexión
Conectividad básica :
La herramienta ahora realiza una prueba básica de conectividad TCP
Compruebe si el puerto 8089 es accesible
Verificar el enrutamiento de la red y los firewalls
Problemas de SSL :
Si ve errores de SSL, intente configurar VERIFY_SSL=false
Comprobar la validez del certificado y la cadena de confianza
Verificar que el nombre de host coincida con el certificado
Problemas de autenticación :
Verificar las credenciales de Splunk
Comprobar los permisos de usuario
Asegúrese de que la cuenta no esté bloqueada
Depuración :
Establezca FASTMCP_LOG_LEVEL=DEBUG para obtener registros detallados
Consulte los registros de conexión para ver si hay mensajes de error específicos
Revisar los mensajes de configuración de SSL
Problemas de conexión SSE :
Verifique que el punto final de SSE sea accesible a través de /sse
Verifique que los encabezados de tipo de contenido sean adecuados
Utilice las herramientas para desarrolladores del navegador para inspeccionar las conexiones SSE
Puede integrar Splunk MCP con Claude Desktop configurándolo para usar el modo SSE o STDIO. Agregue la siguiente configuración a su claude_desktop_config.json :
{
"mcpServers": {
"splunk": {
"command": "poetry",
"env": {
"SPLUNK_HOST": "your_splunk_host",
"SPLUNK_PORT": "8089",
"SPLUNK_USERNAME": "your_username",
"SPLUNK_PASSWORD": "your_password",
"SPLUNK_SCHEME": "https",
"VERIFY_SSL": "false"
},
"args": [
"--directory",
"/path/to/splunk-mcp",
"run",
"python",
"splunk_mcp.py",
"stdio"
]
}
}
}{
"mcpServers": {
"splunk": {
"command": "poetry",
"env": {
"SPLUNK_HOST": "your_splunk_host",
"SPLUNK_PORT": "8089",
"SPLUNK_USERNAME": "your_username",
"SPLUNK_PASSWORD": "your_password",
"SPLUNK_SCHEME": "https",
"VERIFY_SSL": "false",
"FASTMCP_PORT": "8001",
"DEBUG": "true"
},
"args": [
"--directory",
"/path/to/splunk-mcp",
"run",
"python",
"splunk_mcp.py",
"sse"
]
}
}
}Una vez configurado, puede usar lenguaje natural para interactuar con Splunk a través de Claude. Ejemplos:
Lista de índices disponibles:
What Splunk indexes are available?Buscar datos de Splunk:
Search Splunk for failed login attempts in the last 24 hoursObtener el estado del sistema:
Check the health of the Splunk systemAdministrar tiendas KV:
List all KV store collectionsLas herramientas MCP estarán disponibles automáticamente para Claude, lo que le permitirá ejecutar estas operaciones a través de comandos de lenguaje natural.
[Su licencia aquí]
Marco FastMCP
SDK de Splunk para Python
Python-decouple para la gestión de la configuración
SSE Starlette para la implementación de SSE
Get information about the currently authenticated user.
This endpoint retrieves:
- Basic user information (username, real name, email)
- Assigned roles
- Default app settings
- User type
Returns:
Dict[str, Any]: Dictionary containing user information
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
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 clearly describes what information is retrieved (e.g., basic info, roles, settings) and the return type, but lacks details on authentication requirements, error handling, or rate limits. This is adequate but has gaps for a tool with no annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured and front-loaded, starting with the core purpose followed by bullet points for retrieved information and return details. Every sentence adds value without redundancy, making it efficient and easy to parse.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's low complexity (0 parameters, no output schema, no annotations), the description is largely complete, covering purpose, retrieved data, and return type. However, it could benefit from more behavioral context, such as authentication notes, to fully compensate for the lack of annotations.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0 parameters with 100% coverage, so no parameter documentation is needed. The description appropriately does not discuss parameters, focusing instead on the tool's function and output, which aligns with the schema's simplicity.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description explicitly states the verb 'Get' and resource 'information about the currently authenticated user,' making the purpose specific and clear. It distinguishes this tool from siblings like 'list_users' by focusing on the current authenticated user rather than listing all users.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage context by specifying 'currently authenticated user,' suggesting it should be used when user-specific data is needed. However, it does not explicitly state when to use alternatives like 'list_users' or provide exclusions, leaving some ambiguity.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get a list of all indexes and their sourcetypes.
This endpoint performs a search to gather:
- All available indexes
- All sourcetypes within each index
- Event counts for each sourcetype
- Time range information
Returns:
Dict[str, Any]: Dictionary containing:
- indexes: List of all accessible indexes
- sourcetypes: Dictionary mapping indexes to their sourcetypes
- metadata: Additional information about the search
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
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 effectively describes what the tool does (performs a search to gather specific data types) and outlines the return structure in detail. However, it doesn't mention potential limitations like performance impact, authentication requirements, or rate limits, which would be helpful for a search operation.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is perfectly structured: a clear purpose statement followed by bullet points detailing what the search gathers and a formatted Returns section. Every sentence adds value with zero redundancy, making it easy to scan and understand.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter tool with no annotations or output schema, the description provides excellent context about the operation and return values. It explains the search behavior and output structure thoroughly. The only minor gap is lack of information about potential side effects or performance characteristics.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0 parameters with 100% coverage, so no parameter documentation is needed. The description appropriately focuses on output semantics instead, detailing what data is returned. This exceeds the baseline expectation for a zero-parameter tool.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('Get a list') and resources ('all indexes and their sourcetypes'), distinguishing it from siblings like list_indexes (which likely just lists indexes) and get_index_info (which likely provides detailed info about a specific index). The verb+resource combination is precise and unambiguous.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for gathering comprehensive index and sourcetype data with event counts and time ranges, but doesn't explicitly state when to use this tool versus alternatives like list_indexes or get_index_info. There's no guidance on prerequisites, exclusions, or specific scenarios where this tool is preferred over siblings.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get metadata for a specific Splunk index.
Args:
index_name: Name of the index to get metadata for
Returns:
Dictionary containing index metadata
| Name | Required | Description | Default |
|---|---|---|---|
| index_name | Yes |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It states it 'gets metadata' which implies a read-only operation, but doesn't specify permissions required, rate limits, error conditions, or what specific metadata fields are returned. The return format is vaguely described as a 'dictionary', lacking detail on structure or content.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is efficiently structured with a clear purpose statement followed by Args and Returns sections. Every sentence serves a purpose: the first states the tool's function, the second explains the parameter, and the third describes the return. No redundant or verbose language is present.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a single-parameter read operation with no annotations and no output schema, the description provides basic purpose and parameter explanation but lacks sufficient behavioral context. It doesn't detail what metadata is included, error handling, or authentication requirements, leaving gaps for an agent to use the tool effectively without additional context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 0%, but the description compensates by explaining the single parameter 'index_name' as 'Name of the index to get metadata for', adding semantic meaning beyond the schema's title 'Index Name'. However, it doesn't provide format examples, constraints, or validation rules for the index name parameter.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb 'Get' and resource 'metadata for a specific Splunk index', making the purpose unambiguous. It distinguishes from siblings like 'list_indexes' (which lists multiple indexes) by focusing on metadata retrieval for a single specified index. However, it doesn't explicitly contrast with 'get_indexes_and_sourcetypes' which might overlap in functionality.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage when you need metadata for a specific index (vs. listing indexes), but doesn't provide explicit guidance on when to choose this tool over alternatives like 'get_indexes_and_sourcetypes' or 'list_indexes'. No prerequisites, exclusions, or comparative context are mentioned, leaving usage decisions to inference.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get basic Splunk connection information and list available apps (same as health_check but for endpoint consistency)
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
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 states what the tool does (get connection info and list apps) but lacks details on permissions required, rate limits, error conditions, or response format. For a tool with zero annotation coverage, this leaves significant behavioral gaps.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is perfectly concise and front-loaded: the first clause states the core purpose, and the second provides important sibling context. Every sentence earns its place with zero wasted words or redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (0 parameters, no annotations, no output schema), the description is reasonably complete for a basic health/status tool. However, without annotations or output schema, it should ideally mention what specific information is returned or any authentication requirements. The sibling comparison adds useful context.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters with 100% schema description coverage, so the schema already fully documents the parameter situation. The description appropriately doesn't discuss parameters since none exist, which is correct. Baseline for 0 parameters is 4.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Get basic Splunk connection information and list available apps'. It specifies both the verb ('Get') and the resources ('connection information', 'available apps'), making the function unambiguous. However, it doesn't explicitly differentiate from its sibling 'health_check' beyond noting they're functionally equivalent for endpoint consistency.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context by stating this tool is 'same as health_check but for endpoint consistency', which implicitly guides usage toward this tool when endpoint consistency is needed. It doesn't explicitly state when NOT to use it or name alternatives beyond the mentioned sibling, but the comparison offers practical guidance.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get basic Splunk connection information and list available apps
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. While it states what the tool does, it doesn't describe important behavioral aspects like authentication requirements, rate limits, error conditions, or what specific information constitutes 'basic Splunk connection information.' The description is functional but lacks operational context.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence that communicates the complete purpose without any wasted words. It's appropriately sized for a zero-parameter diagnostic tool and front-loads the essential information. Every word earns its place in the description.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given zero parameters and no output schema, the description adequately covers what the tool does but leaves important questions unanswered about return format, error handling, and authentication requirements. For a diagnostic tool that likely returns structured data about system health, more detail about what constitutes 'basic connection information' would be helpful.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has zero parameters with 100% schema description coverage, so the schema already fully documents the parameter situation. The description appropriately doesn't discuss parameters since none exist, which is correct for this case. A baseline of 4 is appropriate when no parameters need documentation.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with specific verbs ('Get', 'list') and resources ('basic Splunk connection information', 'available apps'), making it immediately understandable. It distinguishes this tool from siblings like 'ping' (simple connectivity) or 'list_users' (specific resource listing) by combining connection verification with app inventory.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for checking Splunk connectivity and discovering apps, but doesn't explicitly state when to use this versus alternatives like 'ping' (simpler connectivity check) or 'list_saved_searches' (specific resource listing). No guidance is provided about when NOT to use this tool or what prerequisites might be needed.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Get a list of all available Splunk indexes.
Returns:
Dictionary containing list of indexes
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It states the tool returns a dictionary with a list of indexes, which is basic behavioral information. However, it lacks details on permissions, rate limits, or whether the list is paginated or filtered, which are important for a read operation in a system like Splunk.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is concise with two sentences that directly state the purpose and return value. It's front-loaded with the main action, though it could be slightly more structured by explicitly separating usage notes from return details.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (0 parameters, no output schema, no annotations), the description is adequate but minimal. It covers the basic purpose and return format, but for a tool in a complex system like Splunk, it could benefit from more context on limitations or typical use cases.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters, and schema description coverage is 100%, so no parameter information is needed. The description doesn't add parameter details, but this is acceptable given the lack of parameters, aligning with the baseline for 0 parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with 'Get a list of all available Splunk indexes,' which is a specific verb+resource combination. However, it doesn't differentiate from sibling tools like 'get_indexes_and_sourcetypes' or 'get_index_info,' which might offer overlapping or related functionality.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
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. It doesn't mention sibling tools or contexts where this tool is preferred, such as for a simple list versus more detailed index information available in other tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all KV store collections across apps.
Returns:
List of KV store collections with metadata including app, fields, and accelerated fields
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
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 states the tool lists collections and describes the return format, which is helpful. However, it lacks details on potential side effects, authentication needs, rate limits, or error handling, leaving gaps in behavioral context for a tool that might involve data access across apps.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise with only two sentences, front-loading the purpose and then detailing the return format. Every sentence earns its place by providing essential information without any wasted words or redundancy.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (listing collections across apps) and the absence of annotations and output schema, the description is minimally adequate. It covers the purpose and return format, but for a tool that might involve cross-app data access, more context on permissions, scope, or limitations would enhance completeness.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The input schema has 0 parameters with 100% coverage, so no parameter documentation is needed. The description appropriately focuses on the return values instead, adding value by specifying what metadata is included in the output, which compensates for the lack of an output schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the action ('List all KV store collections') and the resource ('across apps'), which provides a specific verb+resource combination. However, it doesn't explicitly differentiate from sibling tools like 'list_indexes' or 'list_saved_searches', which would require mentioning what makes KV store collections distinct from those other resources.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
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. It doesn't mention prerequisites, context for usage, or compare it to sibling tools like 'get_indexes_and_sourcetypes' or 'list_indexes', leaving the agent without explicit direction on selection criteria.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all saved searches in Splunk
Returns:
List of saved searches with their names, descriptions, and search queries
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries full burden for behavioral disclosure. It states the return format ('List of saved searches with their names, descriptions, and search queries'), which is helpful, but lacks details on permissions, rate limits, pagination, or error handling for a read operation in Splunk.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded with the core purpose in the first sentence and adds return details in a second sentence. It's efficient with zero waste, though slightly brief for a tool with no annotations, which keeps it from a perfect score.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given 0 parameters, 100% schema coverage, and no output schema, the description is minimally adequate. It explains what the tool does and the return format, but for a read operation in Splunk with no annotations, it could benefit from more behavioral context like access requirements or data scope.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters, and schema description coverage is 100%, so no parameter documentation is needed. The description appropriately omits parameter details, earning a high baseline score for not adding unnecessary information.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('List') and resource ('all saved searches in Splunk'), making the purpose unambiguous. However, it doesn't differentiate from sibling tools like 'list_indexes' or 'list_users' beyond specifying the resource type, which prevents a perfect score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. The description doesn't mention prerequisites, limitations, or comparisons to sibling tools like 'search_splunk' or 'get_indexes_and_sourcetypes', leaving the agent without context for selection.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all available MCP tools.
Returns:
List of all available tools with their name, description, and parameters.
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It mentions the return format ('List of all available tools with their name, description, and parameters'), which is helpful. However, it doesn't disclose important behavioral traits like whether this is a read-only operation, if it requires authentication, rate limits, or pagination behavior. The description provides some output information but misses key operational details.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is extremely concise and well-structured. Two sentences efficiently convey the tool's purpose and return format with zero wasted words. It's front-loaded with the core functionality and follows with return details.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's simplicity (0 parameters, no output schema, no annotations), the description is adequate but has gaps. It explains what the tool does and what it returns, but doesn't provide context about when to use it or behavioral constraints. For a tool with no structured metadata, the description should do more to compensate, particularly around usage guidelines and operational behavior.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters with 100% schema description coverage. The description doesn't need to explain parameters since none exist. It appropriately focuses on the tool's function and return value rather than parameter details. The baseline for 0 parameters is 4.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'List all available MCP tools' with a specific verb ('List') and resource ('MCP tools'). It distinguishes from siblings by focusing on tools rather than users, indexes, searches, etc. However, it doesn't explicitly differentiate from similar listing tools like list_indexes or list_users beyond the resource type.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
No guidance is provided on when to use this tool versus alternatives. The description doesn't mention prerequisites, timing considerations, or compare it to other listing tools. It simply states what it does without context about when it's appropriate.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
List all Splunk users (requires admin privileges)
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden. It discloses the admin privilege requirement, which is useful behavioral context. However, it doesn't mention other traits like pagination, rate limits, or return format, leaving gaps in behavioral understanding.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is a single, efficient sentence with zero wasted words. It's front-loaded with the core purpose and includes essential context (admin privileges) without unnecessary elaboration.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a tool with 0 parameters and no output schema, the description covers the basic purpose and permission requirement adequately. However, without annotations or output schema, it lacks details on return format (e.g., list structure, fields) and other behavioral aspects, making it minimally complete.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters, and schema description coverage is 100%, so no parameter documentation is needed. The description appropriately doesn't discuss parameters, earning a baseline 4 for this context.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the verb ('List') and resource ('all Splunk users'), making the purpose unambiguous. However, it doesn't explicitly differentiate from sibling tools like 'current_user' or 'list_indexes', which would require a 5.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description provides clear context with 'requires admin privileges', indicating when this tool is appropriate. It doesn't explicitly state when not to use it or name alternatives like 'current_user' for non-admin scenarios, preventing a score of 5.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Simple ping endpoint to check server availability and get basic server information.
This endpoint provides a lightweight way to:
- Verify the server is running and responsive
- Get basic server information including version and server time
- Check connectivity without making complex API calls
Returns:
Dict[str, Any]: Dictionary containing status and basic server information
| Name | Required | Description | Default |
|---|---|---|---|
No parameters | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
With no annotations provided, the description carries the full burden and does well by disclosing behavioral traits: it's a read-only operation that returns server information, doesn't modify anything, and provides version and server time. It could mention response format details or error behavior, but covers the essential safety profile adequately.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is perfectly structured: opening statement of purpose, bulleted list of specific capabilities, and clear return value description. Every sentence earns its place with zero wasted words, and information is front-loaded effectively.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
For a zero-parameter diagnostic tool with no annotations and no output schema, the description provides complete context: purpose, when to use, what it returns, and how it differs from other tools. The return format description ('Dict[str, Any]') compensates for the missing output schema.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The tool has 0 parameters with 100% schema coverage, so the baseline would be 4. The description appropriately doesn't discuss parameters since none exist, maintaining focus on the tool's purpose and behavior.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose with specific verbs ('check server availability', 'get basic server information') and distinguishes it from siblings by emphasizing it's a 'lightweight way' for connectivity verification rather than complex API operations like search_splunk or list_users.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description explicitly states when to use this tool: 'to check server availability', 'verify the server is running and responsive', and 'check connectivity without making complex API calls'. It clearly distinguishes this from more complex operations available in sibling tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Execute a Splunk search query and return the results.
Args:
search_query: The search query to execute
earliest_time: Start time for the search (default: 24 hours ago)
latest_time: End time for the search (default: now)
max_results: Maximum number of results to return (default: 100)
Returns:
List of search results
| Name | Required | Description | Default |
|---|---|---|---|
| search_query | Yes | ||
| earliest_time | No | -24h | |
| latest_time | No | now | |
| max_results | No |
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 states the tool executes a search and returns results, but lacks details on permissions required, rate limits, whether it's read-only or has side effects, response format, or error handling. For a tool with 4 parameters and no annotation coverage, this is a significant gap in transparency.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is well-structured and appropriately sized. It starts with a clear purpose statement, followed by a bulleted list of args with explanations and defaults, and ends with return information. Every sentence earns its place by adding value, with no redundant or verbose content, making it easy to parse and understand.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's complexity (4 parameters, no annotations, no output schema), the description is partially complete. It covers parameter semantics well but lacks behavioral context like permissions or rate limits, and without an output schema, it doesn't detail the structure of the 'List of search results'. This leaves gaps for an AI agent to fully understand tool behavior and outputs.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
The description adds substantial meaning beyond the input schema, which has 0% description coverage. It explains each parameter's purpose: 'search_query' as the query to execute, 'earliest_time' and 'latest_time' as time bounds with defaults, and 'max_results' as a limit. This compensates fully for the schema's lack of descriptions, providing clear semantics for all parameters.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the tool's purpose: 'Execute a Splunk search query and return the results.' This specifies the verb ('execute') and resource ('Splunk search query'), making it understandable. However, it doesn't explicitly differentiate this tool from potential siblings like 'list_saved_searches' or 'get_indexes_and_sourcetypes', which might also involve search-related operations, so it falls short of a perfect score.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
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. It doesn't mention sibling tools like 'list_saved_searches' for retrieving pre-defined searches or 'get_indexes_and_sourcetypes' for exploring data sources, nor does it specify prerequisites or contexts for executing searches. This lack of comparative information limits its utility for an AI agent in selecting the right tool.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
current_userget_index_infoget_indexes_and_sourcetypeshealthhealth_checklist_indexeslist_kvstore_collectionslist_saved_searcheslist_toolslist_userspingsearch_splunkScored across 12 tools
There is significant overlap between several tools, which could cause confusion. For example, 'health' and 'health_check' appear to be duplicates, and 'list_indexes', 'get_indexes_and_sourcetypes', and 'get_index_info' all relate to indexes with unclear boundaries. However, descriptions help differentiate some tools, such as 'search_splunk' being distinct for query execution.
Naming is inconsistent with mixed conventions. Some tools use verb_noun patterns like 'list_indexes' and 'search_splunk', while others use noun-only forms like 'health' and 'ping'. There are also deviations like 'get_indexes_and_sourcetypes' using 'and' in the name, and 'list_tools' is an outlier as it's meta to the server itself.
With 12 tools, the count is reasonable for a Splunk MCP server, covering user management, indexing, searches, and health checks. It's slightly heavy due to redundant tools like 'health' and 'health_check', but overall well-scoped for the domain without being overwhelming.
The toolset covers core Splunk operations like searching, indexing, and user management, but has notable gaps. For example, there are tools to list saved searches but no way to create, update, or delete them, and similar gaps exist for KV store collections and indexes. This could lead to dead ends for agents trying to perform full lifecycle management.
Official Microsoft MCP Server to query Microsoft Entra data using natural language
Query your org's data in natural language — read-only MCP access to SQL, NoSQL, files & warehouses.
Your org's AI agents, tasks, runs, search, and brain files as MCP tools and resources.
Governed data discovery, exact queries, decisions, simulations, and runtime utilities over MCP.
