Karma MCP Server
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| KARMA_URL | Yes | The URL of your Karma instance |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| check_karmaA | Check connection to Karma server |
| list_alertsB | List all active alerts in Karma |
| get_alerts_summaryA | Get a summary of alerts grouped by severity and state |
| list_clustersA | List all available Kubernetes clusters in Karma |
| list_alerts_by_clusterA | List alerts filtered by specific cluster Args: cluster_name: Name of the cluster to filter by (e.g., 'teddy-prod', 'edge-prod') |
| get_alert_detailsC | Get detailed information about a specific alert Args: alert_name: Name of the alert to get details for |
| list_active_alertsA | List only active (non-suppressed) alerts |
| list_suppressed_alertsC | List only suppressed alerts |
| get_alerts_by_stateC | Get alerts filtered by state (active, suppressed, or all) |
| get_alert_details_multi_clusterA | Get detailed information about a specific alert across multiple clusters Args: alert_name: Name of the alert to search for (e.g., 'KubePodCrashLooping') cluster_filter: Optional cluster name filter. If empty, searches all clusters. |
| search_alerts_by_containerA | Search for alerts by container name across multiple clusters Args: container_name: Name of the container to search for cluster_filter: Optional cluster name filter (e.g., 'teddy-prod', 'edge-prod'). If empty, searches all clusters. |
| list_silencesA | List all active silences across clusters or for a specific cluster Args: cluster: Optional cluster name to filter silences (e.g., 'teddy-prod') Returns: Formatted list of active silences with details |
| create_silenceA | Create a new silence for specific alerts Args: cluster: Target cluster name (e.g., 'teddy-prod') alertname: Name of the alert to silence duration: Duration of silence (e.g., '2h', '30m', '1d') comment: Comment explaining why the alert is being silenced matchers: Additional matchers in format 'key=value,key2=value2' (optional) Returns: Silence creation result with silence ID |
| delete_silenceA | Delete (expire) an existing silence Args: silence_id: ID of the silence to delete cluster: Cluster where the silence exists Returns: Deletion result |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| get_alert_analysis_prompt | Comprehensive prompt for alert analysis and troubleshooting |
| get_incident_response_prompt | Guidelines for incident response and escalation |
| get_kubernetes_context_prompt | Kubernetes-specific context for alert interpretation |
| get_business_impact_prompt | Business impact assessment guidelines for alert prioritization |
TDQS
Scored across 14 tools
Each tool has a distinct purpose: health check, silence management, alert listing/filtering by various criteria, cluster listing, and container search. No overlapping ambiguity.
All tools use consistent lowercase underscore-separated verb_noun naming (e.g., create_silence, list_alerts, get_alert_details). No mixed conventions.
14 tools cover all necessary operations for a Karma alert management server: connection check, silence CRUD, alert listing with various filters, and cluster info. Well-scoped.
The tool set covers the full range of Karma operations: checking server health, managing silences (create/delete), listing and filtering alerts by state, cluster, container, and getting details. No obvious gaps.