GTEx MCP defines 5 tools with basic schemas and descriptions. All tools have input schemas with type definitions and most parameters are described. However, descriptions are minimal (10-80 chars, well below the 194-char baseline), lack context on WHEN to use each tool, and omit dependency hints. Parameter descriptions are sparse: 'tissue' in top_expressed_genes lacks format guidance (what is a tissueSiteDetailId?), 'num' lacks explicit range validation in description, and 'gencode_id' vs 'gencode_id_or_symbol' naming is inconsistent. No output schemas documented. Error handling is basic (generic HTTP errors). Tools are read-only (low risk) but lack idempotency hints. Naming follows verb_noun pattern (gene, median_expression) but some names are vague: 'gene' could mean lookup, create, or update, should be 'get_gene'. No tool annotations present.

• Expand tool descriptions to 100-200 chars. Include WHAT the tool does, WHEN to use it, and any prerequisites. E.g., 'Get gene metadata (name, chromosome, transcript count) by Gencode ID or symbol. Use this first to validate a gene exists before calling median_expression or single_tissue_eqtls.'
• Replace example values in parameter descriptions with formal constraints. Change 'GTEx tissueSiteDetailId (e.g. "Liver", "Whole_Blood")' to an enum: {"type": "string", "enum": ["Liver", "Whole_Blood", ...]} or a pattern regex.
• Document output schemas for all tools. Add a 'returns' field to each tool definition describing the response structure. E.g., gene returns {gene_id, symbol, chromosome, transcript_count, ...}.
• Rename 'gene' tool to 'get_gene' for clarity. Ensure all parameter names are consistent (gencode_id, not gencodeId in schema).
• Add readOnlyHint: true to all tool definitions since all are read-only queries. This signals to agents that these tools are safe to call without confirmation.