StackQL MCP server has 12 tools with basic schemas and descriptions, but significant gaps limit production readiness. All tools have descriptions (10-100 chars, mostly adequate), and input schemas are present with type definitions. However, output schemas are entirely undocumented, LLMs cannot predict what fields to expect from responses, breaking tool chaining. Parameter descriptions are minimal (e.g., 'Cloud provider name' lacks format/constraint guidance). No error handling guidance, no recovery paths, and no tool annotations (readOnlyHint/destructiveHint). The exec_query and pull_provider tools are destructive but lack confirmation patterns or dry-run support. Tool naming is verb-first and clear, but composition could be tighter, list_providers and list_registry appear redundant.

• Document output schemas for all 12 tools. Specify field names, types, and whether results are paginated. Example: exec_query should return {results: [{column: string, value: any}], rowCount: number, executionTime: number}.
• Add tool annotations: mark exec_query, pull_provider, reload_credentials with destructiveHint=true; mark all read-only tools with readOnlyHint=true; mark idempotent tools with idempotentHint=true.
• Consolidate list_providers and list_registry into a single canonical tool. If they differ, rename one to clarify the distinction (e.g., list_installed_providers vs list_available_providers).
• Add enums to provider, service, resource, and method parameters. If the registry is dynamic, document the discovery pattern: 'Call list_providers first to see valid provider names.'
• Implement confirmation pattern for exec_query: add optional dry_run=true parameter that returns the query plan without executing, allowing agents to preview before committing.