Chat2DB CLI exposes 26 tools with minimal schema rigor. All tools have basic descriptions (10-50 chars), but most lack depth. Input schemas are present but sparse: many parameters lack type constraints, enums, or validation guidance. No output schemas documented. Tool names follow verb_noun convention (mcp_status, sql_query), but descriptions are generic and fail to explain WHEN to use each tool or what happens on failure. Error handling is absent, no recovery guidance, no categorization of retryable vs fatal errors. The server is STDIO-only, which is a hard transport limitation.

• Add output schemas to all tools. Document return types, field names, and structure. Example: mcp_status should return {status: string, edition: string, version: string}.
• Expand descriptions to 100-200 chars. Include WHEN to use the tool, prerequisites, and what happens on success/failure. Example: 'Enable MCP server for the specified edition (community, pro, or local). Required before calling mcp_call. Returns success status or error if already enabled.'
• Add enum constraints to 'edition' parameter across all tools: {type: 'string', enum: ['community', 'pro', 'local'], description: 'Edition to operate on...'}.
• Implement error handling with recovery guidance. For destructive tools (uninstall, skill_uninstall), add a dry_run parameter and return detailed error messages like 'Skill not found. Available skills: [list]'.
• Document the 'arguments' parameter in mcp_call with a schema or example. Specify which tools accept which argument shapes to guide LLM composition.