Three tools with complete schemas and descriptions, but significant gaps in error handling, parameter validation, and output documentation. Tool names follow verb_noun convention (uk_market_search, uk_land_registry, uk_hpi). Descriptions are present (100-150 chars) but lack actionable guidance on when to use each tool vs. alternatives. Parameters have types and descriptions, but lack constraints (enums, ranges, patterns). Output is JSON strings rather than structured objects. No error recovery guidance. Annotations present but incomplete (missing destructiveHint/idempotentHint where applicable).

• Return structured Pydantic models instead of JSON strings. Define output schemas (e.g. class MarketSearchResult with fields: total_listings: int, stc_rate_pct: float, listings: List[Listing]) and document them in tool docstrings.
• Add input validation with clear error messages. E.g., 'radius must be 0.1 - 50 miles; got 10000' instead of raw exception. Validate postcode format before calling Rightmove API.
• Add parameter constraints: radius (0.1 - 50, default 1), max_price (1 - 10000000, default 300000), months (1 - 120, default 12). Use Field(ge=, le=) in Pydantic.
• Enhance tool descriptions with selection guidance: 'Use uk_market_search for current listings and market sentiment; use uk_land_registry for historical sales data and price trends; use uk_hpi for regional price indices.'
• Implement error recovery patterns. When postcode lookup fails, return 'Postcode not found. Try a nearby postcode or use a town name instead.' When SPARQL times out, suggest retrying or using a smaller date range.