Four tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. Tool names follow verb_noun convention (setup, list-profiles, test-connection, delete-profile), which is good. However, descriptions are minimal (10-50 chars), parameter descriptions lack detail on constraints and formats, and output schemas are not documented. No error handling guidance visible. The setup tool exposes sensitive parameters (api_key, api_secret) which violates secret-injection pattern. No tool annotations (readOnlyHint, destructiveHint) despite clear risk levels.

• Move api_key and api_secret out of tool parameters. Use keyring library (already in dependencies) to store credentials server-side. setup tool should only accept profile name, url, and verify_ssl; prompt for secrets interactively or read from environment.
• Document output schemas for all four tools. E.g., list-profiles should return {profiles: [{name: string, url: string, created_at: ISO8601}], total: number}. setup should return {profile: string, status: 'success'|'failed', message: string}.
• Add enum constraints to profile parameter. Instead of free-form string, define allowed profile names or document the naming convention (alphanumeric, hyphens, max 50 chars).
• Add tool annotations to tool definitions. Mark delete-profile with destructiveHint=true, list-profiles with readOnlyHint=true, setup with destructiveHint=true (modifies config). This requires fastmcp support for annotations in tool registration.
• Add error handling examples in descriptions. E.g., 'If connection fails, verify URL is reachable and credentials are correct. Call test-connection to diagnose.' This guides agent recovery.

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

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