Three tools with minimal documentation. All have basic descriptions (40-80 chars, below the 194-char baseline), but lack parameter descriptions entirely. Input schemas present but sparse: only 1-2 string parameters per tool, no enums, no constraints, no output schemas documented. No error handling guidance. Tool names are verb-noun (good), but descriptions do not explain WHEN to use each tool or what happens on failure. Composition is reasonable (list → get → run), but missing chaining IDs and pagination.

• Add 50 - 150 character descriptions to each parameter explaining its type, format, and how to obtain it. E.g., 'dataflow_id: The UUID or slug of the Coupler dataflow (e.g., "my-etl-flow" or "550e8400-e29b-41d4-a716-446655440000")'.
• Document output schemas for all three tools. E.g., list-datasets should return [{snapshot_id, last_success_timestamp, row_count}] so get-data knows which snapshot_id to pass.
• Expand tool descriptions to 100 - 200 chars. Include WHEN to call (e.g., 'Call this after list-datasets to fetch actual data from a snapshot') and what happens on error (e.g., 'Returns error if snapshot is stale or query times out after 30s').
• Add error handling guidance: 'If query fails, check syntax and retry. If snapshot is stale, call run-dataflow first to refresh.' Include timeout and row-limit constraints in get-data description.
• Consider adding optional parameters to get-data: limit (default 100, max 10000) and timeout_seconds (default 30, max 300) to prevent runaway queries and context explosion.