feat: add Loki log query support to lab-monitoring
Add 3 opt-in Loki tools (query_logs, list_labels, list_label_values) that are registered when LOKI_URL is configured. Includes Loki HTTP client, CLI commands (logs, labels), NixOS module option, formatting, and tests. Co-Authored-By: Claude Opus 4.5 <noreply@anthropic.com>
This commit is contained in:
@@ -83,7 +83,7 @@ This ensures package information matches the nixpkgs version the project actuall
|
||||
func DefaultMonitoringConfig() ServerConfig {
|
||||
return ServerConfig{
|
||||
Name: "lab-monitoring",
|
||||
Version: "0.1.0",
|
||||
Version: "0.2.0",
|
||||
Mode: ModeCustom,
|
||||
Instructions: `Lab Monitoring MCP Server - Query Prometheus metrics and Alertmanager alerts.
|
||||
|
||||
@@ -92,8 +92,9 @@ Tools for querying your monitoring stack:
|
||||
- List and inspect alerts from Alertmanager
|
||||
- View scrape target health status
|
||||
- Manage alert silences
|
||||
- Query logs via LogQL (when Loki is configured)
|
||||
|
||||
All queries are executed against live Prometheus and Alertmanager HTTP APIs.`,
|
||||
All queries are executed against live Prometheus, Alertmanager, and Loki HTTP APIs.`,
|
||||
}
|
||||
}
|
||||
|
||||
|
||||
Reference in New Issue
Block a user