Bridge your AI assistant to PostgreSQL databases — query, execute, migrate, and analyze without cloud dependencies
DB Bridge has 6 tools with clear verb-noun naming (query, execute, list_tables, describe_table, run_migration, explain_analyze). All tools have descriptions (avg 85 chars, within baseline 34-392). Input schemas are present with typed parameters. However, parameter descriptions are minimal or missing context about constraints, formats, and dependencies. Output schemas are not explicitly documented in the code. Error handling is present but lacks recovery guidance. The run_migration tool (IRREVERSIBLE risk) lacks confirmation/dry-run pattern. No tool annotations (readOnlyHint/destructiveHint) despite clear risk levels.
Describe a table's structure including columns, indexes, and constraints.
Execute a parameterized INSERT, UPDATE, or DELETE statement.
Get query execution plan with costs using EXPLAIN ANALYZE.
List all tables in a schema.
Execute a parameterized SELECT query and return rows.
Run a database migration (DDL statements) in a transaction.
run_migration lacks confirmation/dry-run pattern for irreversible DDL operations. No mechanism to preview changes or require explicit approval before executing migrations.
No tool annotations (readOnlyHint, destructiveHint, idempotentHint) despite clear risk levels (READ_ONLY, WRITE, IRREVERSIBLE). LLMs cannot infer safety properties from tool names alone.
Parameter descriptions lack constraint details. 'schema' param has no default behavior documented; 'sql' params lack format guidance (e.g., parameterized vs raw SQL). No validation rules stated.
Inferred effective spec: 2026-07-28+.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-23 | D | 57 | 2026-07-28+ | v2 |
Output schemas not documented in code. Return types are inferred from docstrings ('Dictionary with rows, row_count, columns...') but formal schema definitions are absent. LLMs cannot reliably parse response structure.
Error handling lacks recovery guidance. No indication of retryable vs fatal errors, or what the LLM should do next (e.g., 'Try with a different schema' or 'Check connection parameters').