Single tool 'list_tenants' with empty input schema and minimal description. No parameter documentation, no output schema specification, no error handling guidance. Tool definition is visible in server.py but lacks the structured metadata required for LLM-driven agent composition. Description is 34 chars (below 50-char baseline for clarity). No input validation, no pagination support despite multi-tenant context suggesting potential for large result sets.

• Expand tool description to 100-150 chars: 'List all configured Prometheus tenants with their URLs and authentication status. Call this first to discover available monitoring targets before querying metrics.'
• Add output schema documenting returned fields: {tenants: [{name: string, url: string, org_id?: string, auth_type?: 'basic'|'token'|'none'}], total: number, default_tenant: string}
• Add optional parameters: limit (1-100, default 20), offset (default 0) to support pagination for large multi-tenant deployments.
• Document error cases: 'If no tenants configured, returns error 'No tenants found. Configure PROMETHEUS_TENANTS env var.' If config is malformed, returns 'Invalid tenant configuration: {details}'.'
• Add per-tenant validation status to response so agents can detect misconfigured tenants without making separate calls.

Inferred effective spec: <=2025-11-25.

• Relies on Logging (deprecated) - log to stderr or use OpenTelemetry