Two tools with basic schemas and descriptions, but significant gaps in parameter documentation and output schema clarity. get_trust_preview has empty input schema ({}), which violates schema requirements. get_agent_trust has proper Zod schema with descriptions for both parameters, but lacks documented output structure. Descriptions are adequate (100-150 chars) but lack actionable context for LLM selection. No error handling guidance, no pagination, no output schema documentation. Tool names follow verb_noun pattern correctly.

• Add explicit output schema documentation for both tools. Example: get_trust_preview returns {score: number 0-100, components: {breadth, recency, value, volume}, timestamp: ISO8601}.
• Wrap axios calls in try-catch. Translate HTTP errors to actionable messages: 'Agent not found on {chain}. Verify agent ID and chain are correct.' instead of raw 500 errors.
• Expand tool descriptions to 100-150 chars with functional context: 'Get a free sample ERC-8004 trust score to understand the scoring model. No wallet required. Returns score 0-100 with component breakdown.'
• Document agent parameter format: 'ERC-8004 agent ID as uint256 (e.g., 6976) or hex string (e.g., 0x1b40). Numeric format preferred.'
• Add pagination/limit guidance if API supports it. If results are always small, state 'Returns single agent record' to set LLM expectations.