SUS healthcare MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| SUS_USAGE_DB | No | Path to the usage database file (e.g., /tmp/sus-usage.db). Used for tracking usage. | |
| SUS_DEEPER_DATA | No | Directory containing reviewed serving artifacts for deeper tools: surface-intervals.json, metadata.json. | |
| SUS_STAGED_DATA | Yes | Directory containing reviewed serving artifacts for registry tools: capacity.parquet, metadata.json. | |
| SUS_CLINICAL_DATA | No | Directory containing reviewed serving artifacts for clinical tools: clinical.parquet, clinical-metadata.json. | |
| SUS_ENABLE_DEEPER | No | Set to 'true' to enable the additional fixed deeper tools. | false |
| SUS_KNOWN_CARE_DATA | No | Directory containing reviewed serving artifacts for known care tools: known-care.json, known-care-metadata.json. | |
| SUS_ENABLE_KNOWN_CARE | No | Set to 'true' to enable the additional fixed known care tools. | false |
| SUS_ENABLE_CLINICAL_2024 | No | Set to 'true' to enable the additional fixed clinical 2024 tools. | false |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| dataset_infoA | Call first: coverage, vintage, units, privacy release and audit status. |
| geographic_signalsC | Fixed annual municipal diagnosis activity intervals, not patients. Small cells omitted. |
| facility_capacityC | Registered capacity by facility; inpatient/complementary bed types remain separate. |
| dataset_coverageD | Inventory and publisher gates, never treat unpublished months as zero. |
| fixed_proxy_activityD | Reviewed fixed ecological proxy cells. No flexible CID/procedure intersections. |
| ecological_hitsC | Pre-reviewed fixed multi-signal ecological hits, not patient case finding. |
| protocol_billing_associationC | Dated protocol/billing association only; never diagnosis or patient eligibility. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
No prompts | |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 7 tools
dataset_info and dataset_coverage both describe coverage/availability metadata, making them easy to confuse. geographic_signals, fixed_proxy_activity, and ecological_hits all return fixed ecological/proxy signals with subtle jargon-based distinctions that an agent may not reliably parse.
All names are snake_case noun phrases with no camelCase or mixed-style violations, so the surface is readable. However, there is no consistent verb_noun action pattern and prefix conventions vary (dataset_* versus standalone noun phrases).
Seven tools is well within the ideal 3–15 range for a domain-specific data server. Each tool maps to a distinct data product or metadata concern, so the count feels well-scoped.
The surface covers metadata, coverage, geographic signals, facility capacity, proxy activity, ecological hits, and protocol/billing association, which is broad for ecological SUS data. A minor gap is the absence of an explicit data dictionary/variable lookup or flexible query/filter tool, but core read-only data products are present.