dbboard-mcp has 8 tools with basic naming and descriptions, but critical gaps in schema documentation and parameter descriptions. Tool names follow verb_noun convention (list_*, describe_*, query, export_*, ai_*, capture_*), which is good. However, most tools lack visible input schemas in the source code provided. Descriptions are present but generic (10-50 chars for most), falling below the 50-200 char baseline for LLM-optimized descriptions. The query and ai_* tools accept complex parameters (connection, sql, schema) but parameter descriptions are minimal. No output schemas are documented. Error handling is not evident in the code samples. Security concern: export_connections accepts a passphrase parameter, which could be logged. Overall, this is a D-grade server with functional naming but insufficient schema and description depth for production agent use.

• Expand all tool descriptions to 50-200 characters. Include WHAT the tool does, WHEN to use it, and what it returns. Example: 'List all configured database connections from connections.toml. Use this first to discover available databases before querying. Returns connection names and types.'
• Add explicit input schemas for all tools with type definitions and descriptions for every parameter. For query tool: document that sql must be read-only, provide regex pattern or validation rules, and explain how the tool enforces this.
• Document output schemas for all tools. Specify field names, types, and whether results are paginated. Example for list_tables: '{tables: [{name: string, row_count: number, columns: number}], total: number}'
• Move passphrase from export_connections parameter to server-side configuration (environment variable or vault). Never expose secrets as tool parameters.
• Add error handling guidance. Document what errors each tool can return (e.g., 'connection not found', 'invalid SQL syntax') and what the LLM should do next (retry, ask user, try alternative tool).