This server has 11 tools with significant definition gaps. Most tools lack proper input schema visibility in the provided source code. Descriptions vary widely, some are detailed (e.g., build_dopaminergic_reference_network_tool at ~280 chars), others are minimal (e.g., get_organisms at ~35 chars, below the 50-char baseline). Parameter descriptions are present but inconsistent in quality. No output schemas are documented. Error handling and recovery guidance are absent. The server uses HTTP transport (fastmcp) which is positive, but definition quality is well below production baseline.

• Consolidate duplicate tools: merge 'build_dopaminergic_reference_network_tool' and 'Build Dopaminergic Network' into a single tool with consistent snake_case naming (e.g., 'build_dopaminergic_network'). Remove Title Case variants.
• Expand minimal descriptions to 50-200 chars. For get_organisms, change from 'Get list of supported organisms' to 'Retrieve all supported organisms for protein interaction queries. Call this first to validate organism IDs before search_interactions. Returns organism_id, organism_name, and common_name.'
• Document output schemas for all tools. Add a 'Returns' section to each tool description specifying field names, types, and meaning. E.g., search_interactions returns [{interaction_id: string, protein_a: string, protein_b: string, confidence: float, interaction_type: string}].
• Add error handling guidance. For each tool, document failure modes and recovery steps. E.g., 'If resolve_protein_entity returns not_found, try batch_resolve_proteins with partial name match or search_interactions to discover related proteins.'
• Convert free-form string parameters to enums where applicable. discovery_mode should be enum ['minimal', 'standard', 'comprehensive', 'hypothesis_free']. workflow_type should be enum. This prevents LLM hallucination of invalid values.