Model Context Protocol (MCP) integration for AI-ready FHIR server capabilities. Provides AI agents with tools to discover FHIR server capabilities, manage resources, and query healthcare data.
Single tool (fhir_discover) with moderately detailed schema and description. Naming is acceptable but generic. Parameter descriptions are present with enums for topic selection. Output schema is not documented in the provided code. The tool itself is read-only and low-risk, but the server lacks multiple tools that would demonstrate composition patterns. Error handling and recovery guidance are not visible in the source excerpt.
Discover FHIR server capabilities including resources, search parameters, and operations
No output schema documentation. The tool description does not specify what fields are returned, their types, or structure. LLMs cannot plan downstream actions without knowing the response format.
Generic tool name 'fhir_discover' lacks a clear verb-noun pattern. 'list_fhir_resources', 'get_fhir_capabilities', or 'search_fhir_metadata' would more explicitly signal intent.
Parameter 'resourceType' description is sparse: 'Optional: Filter by resource type'. Does not explain format (e.g., 'FHIR resource type like Patient, Observation'), valid examples, or how filtering works across different topic values.
No error handling or recovery guidance visible. Tool description does not explain failure modes (e.g., invalid FHIR version, unreachable server) or what the agent should do if discovery fails.
| Scored | Grade | Overall | Spec posture | Rubric |
|---|---|---|---|---|
| 2026-09-22 | D | 55 | 2026-07-28+ | v2 |
| 2026-03-09 | F | 25 | - | v1 |
Only one tool provided. A production FHIR agent would need tools to search resources, fetch specific records, create/update data, validate against profiles, and handle transactions. Composition patterns cannot be evaluated.