GTM Data MCP
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| UTM_BUILDER_URL | No | URL of the optional UTM Builder API. Must be used together with UTM_BUILDER_TOKEN. | |
| GTM_CATALOG_PATH | No | Path to a local catalog JSON file for database-free evaluation (e.g. ./data/catalog.json). | |
| UTM_BUILDER_TOKEN | No | Token for the optional UTM Builder API. Must be used together with UTM_BUILDER_URL. | |
| GTM_MCP_BEARER_TOKEN | No | Bearer token required for remote HTTP access to the MCP endpoint (e.g. via /api/mcp). |
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": true
} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| gtm_module_statusA | Describe the enabled catalog backend, UTM integration availability, and restricted-data policy. |
| gtm_search_catalogA | Search people, teams, agencies, vendors, systems, accounts, integrations, definitions, measurement assets, runbooks, policies, and reports. |
| gtm_get_recordA | Get one governed record by ID or key with active relationships. |
| gtm_resolve_ownershipC | Resolve owners, operators, approvers, backups, agencies, vendors, and escalation contacts. |
| gtm_get_personnel_mapA | Return people, teams, agencies, and vendors with active responsibility relationships. |
| gtm_get_account_contextC | Find platform account IDs, contacts, APIs, owners, agencies/vendors, integrations, runbooks, and systems. |
| gtm_get_measurement_inventoryB | Return measurement assets, systems, integrations, reports, ownership, and lineage. |
| gtm_trace_lineageA | Trace governed upstream and downstream relationships up to four levels. |
| gtm_get_data_definitionC | Find governed business-term and technical-field definitions. |
| gtm_find_runbooksC | Find active operating, incident, escalation, and recovery runbooks. |
| gtm_check_readinessB | Check lifecycle, verification, ownership, runbook linkage, and pending source updates. |
| gtm_list_source_updatesA | List reviewable changes detected by source reconciliation; this tool cannot approve or apply them. |
| gtm_list_bulk_templatesB | List governed mass-change templates, constraints, documentation, and verification state. |
| gtm_generate_bulk_templateB | Generate safe CSV headers/examples; draft templates require verification against a current platform export. |
| gtm_validate_bulk_changeB | Validate columns, required values, allowed values, and row limits without uploading or changing a platform. |
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 15 tools
Several read-oriented tools overlap in scope: resolve_ownership and get_personnel_map both return responsibility relationships, and get_measurement_inventory, get_account_context, and get_record all offer relationship-rich snapshots. The descriptions generally clarify the primary focus of each tool, but an agent could still struggle to pick between them.
Tool names are overwhelmingly snake_case with a consistent gtm_ prefix and verb-first structure, e.g., get_, list_, search_, generate_, validate_. The only notable deviation is gtm_module_status, which uses a noun phrase rather than a verb_noun pattern, though it remains readable.
Fifteen tools is within a reasonable range for a broad data governance domain, and most tools serve a distinct read or validation purpose. However, the count feels slightly high because some tools could be consolidated or aligned more tightly around core entities.
The server is strong on read, search, lineage, readiness, and bulk-change validation coverage, so most query-oriented workflows are supported. However, there are no lifecycle or mutation tools: source updates can be listed but not approved/applied, records cannot be created or updated, and bulk changes can be generated and validated but not submitted. This leaves notable workflow dead ends, even if the read-only boundary is intentional.