MCP server for the Wazuh SIEM/XDR platform - query agents, alerts, rules, and decoders
The server demonstrates solid definition quality across 27 tools with consistent naming, descriptions, and input schemas. All tools follow verb_noun convention (list_*, get_*, etc.). Tool descriptions are generally adequate (100-250 chars), and parameter schemas are well-defined with proper types and enums where applicable. However, output schemas are not explicitly documented in the provided source, and some parameter descriptions could be more detailed regarding constraints and dependencies. Error handling guidance is present but generic. The lack of visible output schema documentation is the primary gap preventing a higher score.
Get detailed information about a specific Wazuh agent by ID
Get operating system information for a Wazuh agent
Get system statistics (CPU, memory, disk) for a specific Wazuh agent
Get a specific Wazuh alert by ID
Query Wazuh alerts with optional filtering and sorting (requires Wazuh Indexer)
Get details of a specific Wazuh decoder by name
Get diagnostic information for Wazuh components
Output schemas not explicitly documented in source code. While input schemas are well-defined, return types and response field structures are not visible in the provided code samples, making it difficult for LLMs to predict what data will be available after each tool call.
Parameter descriptions for 'sort' fields are generic ('Sort field with direction prefix') without specifying valid sort fields or exact enum values. LLMs cannot determine which fields are sortable without explicit enumeration.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | C | 60 | 2025-06-18+ | v2 |
Get details about a specific Wazuh agent group
Get Wazuh manager status and configuration information
Get details of a specific Wazuh rule by ID
Get details of a specific SCA policy for an agent
Get details of a specific vulnerability by ID
Get Wazuh manager version and build information
List network interfaces on a Wazuh agent
List installed packages on a Wazuh agent
List open network ports on a Wazuh agent
List running processes on a Wazuh agent
List all Wazuh agents with optional status filtering
List Wazuh log decoders with optional filtering
List Wazuh agent groups with optional filtering
List Wazuh manager logs with optional filtering
List rootkit and suspicious file check results for a Wazuh agent
List Wazuh detection rules with optional filtering
List SCA policy checks for an agent and policy
List Security Configuration Assessment (SCA) policies for an agent
List file integrity monitoring (FIM) results for a Wazuh agent
List detected vulnerabilities with optional filtering (requires Wazuh Indexer)
Several tools have optional parameters (agent_id in get_alerts, search/level in list_rules) whose mutual dependencies are not documented. When multiple optional filters exist, LLMs need guidance on valid combinations and defaults.
Tools like list_agents and list_vulnerabilities return paginated results but descriptions do not document the default limit (10), max limits, or whether next_cursor/total_count fields are included in responses for chaining.
get_alerts and list_vulnerabilities both require Wazuh Indexer but this dependency is mentioned in the description rather than enforced via parameter validation or clear error guidance. LLMs may call these tools without knowing a prerequisite.
No error categorization or recovery guidance visible in tool definitions. Tools document they are READ_ONLY but do not provide actionable error messages for common failure modes (invalid agent_id, missing Indexer, timeout, etc.).