cxg-census-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| CXG_CENSUS_MCP_CACHE_DIR | No | Disk cache root | platformdirs default |
| CXG_CENSUS_MCP_LOG_LEVEL | No | stdlib log level | WARNING |
| CXG_CENSUS_MCP_MOCK_MODE | No | If 1, never opens a real Census handle | 0 |
| CXG_CENSUS_MCP_CENSUS_VERSION | No | Census release to pin | stable |
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 |
|---|---|
| census_summaryC | Pinned Census summary: cells, schema, build date. |
| get_census_versionsA | List available Census versions visible to this server. |
| list_available_valuesC | Distinct values for an obs column (discovery before resolve_term). |
| resolve_termC | Resolve text or CURIE to one term; typed refusal if ambiguous. |
| expand_termC | Expand a CURIE to descendants/ancestors filtered to Census presence. |
| term_definitionA | Return label, definition, synonyms, IRI for a CURIE. |
| count_cellsC | Count cells (filters + optional group_by); tier 0. |
| list_datasetsC | List datasets matching a structured filter, sorted by cell count. |
| gene_coverageC | Report whether a list of Ensembl gene IDs is present in Census var. |
| preview_obsC | Small obs slice + column cardinality (before a big scan). |
| aggregate_expressionC | Per-gene expression stats by group; caps → export_snippet if over. |
| export_snippetC | Runnable Python for a stored plan; pass call_id from a prior response. |
| get_server_limitsB | Report cap configuration so clients can size their requests. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| census_workflow | Default workflow guidance for Census exploration. |
| disambiguation | How to handle TERM_AMBIGUOUS responses. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
| Census schema overview | Census schema overview |
| Ontology resolution rules | Ontology resolution rules |
| Server limitations and non-goals | Server limitations and non-goals |
| Recommended agent workflow | Recommended agent workflow |
| Structured error model | Structured error model |
| Progress and cancellation | Progress and cancellation |
TDQS
Scored across 13 tools
Each tool targets a distinct operation: aggregation, counting, term resolution, metadata, etc. No two tools overlap in purpose, and descriptions clearly differentiate them.
All tool names follow a consistent snake_case pattern with descriptive verbs or nouns (e.g., aggregate_expression, get_census_versions, resolve_term). No mixing of conventions.
13 tools is well-scoped for a census data server, covering query, metadata, term expansion, and export without being excessive or insufficient.
The toolset covers core census operations: aggregation, counting, term resolution, dataset listing, limits, versions, and export. No obvious gaps for the server's purpose.