Two tools with basic descriptions and minimal parameter documentation. get_schema has a reasonable description (127 chars) but lacks parameter details. query_data description is incomplete (cut off mid-sentence in source). Both tools have input schemas with types but parameter descriptions are sparse or missing. No output schemas documented. No error handling guidance. No tool annotations. Naming is clear (verb_noun pattern) but parameter naming could be more explicit (e.g., 'table_name' is good, but 'sql_query' lacks format/constraint hints).

• Complete query_data description: 'Executes a read-only SQL SELECT query against the connected database. Returns results as JSON array of objects. Supports pagination via LIMIT/OFFSET. Max 1000 rows per call.'
• Add output schema documentation: get_schema should return {tables: [{name, columns: [{name, type, nullable}]}]} or similar; query_data should return {rows: [...], row_count: number}.
• Add parameter constraints to sql_query: 'Must be a SELECT statement only. No INSERT, UPDATE, DELETE, or DDL. Max 5000 chars. Results capped at 1000 rows.'
• Add error handling examples: 'If database unavailable, returns {error: "Database connection failed. Check DATABASE_URL env var."}. If query invalid, returns {error: "SQL syntax error: ..."}.'
• Add tool annotations to schema: mark both tools with readOnlyHint=true and idempotentHint=true to signal safety to LLM.