QueryPilot exposes 4 database-focused tools with basic descriptions and minimal parameter documentation. All tools have names starting with action verbs (ask_, search_, validate_, execute_) and descriptions present, but descriptions are brief (10-50 chars) and lack context on when to use each tool vs. alternatives. Input schemas are visible but sparse: each tool accepts only 1-2 string parameters with minimal type information. No output schemas are documented in the source. Error handling is not evident from the code sample. The tools appear well-intentioned for SQL safety but lack the depth of documentation and schema rigor expected in production-grade agent tools.

• Expand tool descriptions to 150-250 chars. For ask_database: 'Ask a natural-language question about the database. QueryPilot translates it to safe SQL, validates against policies, and returns results. Use this for exploratory queries when you don't know the schema.'
• Document output schemas for each tool. E.g., ask_database returns {result: string, sql_used: string, rows_affected: number, execution_time_ms: number}. search_schema returns [{table: string, column: string, type: string, description: string}].
• Add parameter descriptions. E.g., 'question' in ask_database: 'Natural-language question about the database (e.g., "Which customers have ARR > $100k?"). Keep under 500 chars.'
• Document the intended workflow: 'Typical flow: (1) search_schema to discover tables, (2) ask_database or validate_sql to explore, (3) execute_sql for final results. Each step is safe and read-only.'
• Add error handling guidance. E.g., 'If validation fails, the response includes the policy violation. Adjust the SQL and retry. If execution times out, try a simpler query or add LIMIT clauses.'