Four tools with complete input schemas and descriptions. Naming follows verb_noun pattern (scan_*, list_*, job_*). Descriptions are substantive (100-200 chars), explaining WHAT and WHEN. Parameters are well-typed with Zod validation and mostly documented. However, output schemas are inferred from code (marketToJson, walletsToJson, jobSummary) rather than explicitly declared in tool registration. Error handling is minimal, no recovery guidance or actionable error messages visible. No tool annotations (readOnlyHint, destructiveHint). Missing pagination guidance for large result sets.

• Add explicit output schema documentation to each tool. Example: 'Returns array of {collection, name, salesCount, volumeEth, pnlAthEth, ...}' in tool description or via structured schema field.
• Add readOnlyHint: true annotation to all four tools in tool registration to signal safety to agents.
• Enhance error messages with recovery guidance. E.g., 'Chain not found. Try list_chains() to see available options.' or 'Job not found. Check job_status() with no jobId to list active jobs.'
• Clarify 'async' parameter: 'If estimated getLogs calls exceed 40, returns jobId immediately; otherwise blocks. Set async=true to force background, async=false to always block.'
• Document result limits in descriptions: 'Returns top 20 collections by default (configurable via top parameter, max 100). Use sortBy to rank by different metrics.'