Odgs MCP server
{
"mcpServers": {
"metricprovenance-odgs": {
"command": "uvx",
"args": [
"odgs-mcp-server"
],
"env": {
"ODGS_API_KEY": "<odgs_api_key>",
"ODGS_REGISTRY_URL": "https://..."
}
}
}
}{
"mcpServers": {
"metricprovenance-odgs": {
"command": "uvx",
"args": [
"odgs-mcp-server"
],
"env": {
"ODGS_API_KEY": "<odgs_api_key>",
"ODGS_REGISTRY_URL": "https://..."
}
}
}
}{
"mcpServers": {
"metricprovenance-odgs": {
"command": "uvx",
"args": [
"odgs-mcp-server"
],
"env": {
"ODGS_API_KEY": "<odgs_api_key>",
"ODGS_REGISTRY_URL": "https://..."
}
}
}
}{
"servers": {
"metricprovenance-odgs": {
"command": "uvx",
"args": [
"odgs-mcp-server"
],
"env": {
"ODGS_API_KEY": "${input:odgs_api_key}",
"ODGS_REGISTRY_URL": "${input:odgs_registry_url}"
}
}
},
"inputs": [
{
"id": "odgs_api_key",
"type": "promptString",
"description": "Odgs api key",
"password": true
},
{
"id": "odgs_registry_url",
"type": "promptString",
"description": "Odgs registry url",
"password": true
}
]
}{
"mcpServers": {
"metricprovenance-odgs": {
"command": "uvx",
"args": [
"odgs-mcp-server"
],
"env": {
"ODGS_API_KEY": "<odgs_api_key>",
"ODGS_REGISTRY_URL": "https://..."
}
}
}
}file .mcp.json (in your project root)file claude_desktop_config.jsonfile ~/.cursor/mcp.jsonfile .vscode/mcp.jsonfile ~/.codeium/windsurf/mcp_config.jsoninstall uvx odgs-mcp-serverversion 0.3.3
What it does
ODGS MCP Server provides runtime validation of data operations against governance rules, producing cryptographically signed audit records (S-Certs). It integrates DAMA DMBOK maturity assessment and regulatory compliance checking for AI systems, particularly those subject to EU AI Act requirements.
- Validate data payloads against governance rules at runtime and generate signed audit certificates for regulatory submission.
- Assess your organisation's data governance maturity across eight DAMA DMBOK pillars with gap analysis.
- Detect definitional drift and semantic conflicts between governance artefacts across versions (with Professional tier).
Tools this server exposes
| Tool | Does |
|---|---|
| governance_score | Assess data governance maturity on a 0–100 scale using 8-pillar DAMA DMBOK methodology with per-pillar gap analysis. |
| validate_payload | Evaluate a data payload against sovereign governance rules and return a signed S-Cert record. |
| conformance_check | Check whether a data operation or artefact conforms to established governance rules. |
| list_packs | List available governance rule packs and certified sovereign packs. |
| compile_regulation | Convert legislative text (EU AI Act, DORA, GDPR) into validated ODGS rule JSON (Professional tier). |
| check_drift | Identify definitional and semantic drift between versions of governance artefacts (Professional tier). |
| detect_conflicts | Detect conflicts between governance rules (Professional tier). |
| narrate_audit | Generate human-readable narration of audit trails and S-Cert records (Professional tier). |
| discover_bindings | Map governance rules to data catalog entities (Professional tier). |
| harvest_sovereign_rules | Extract and mint sovereign rules from Databricks, Snowflake, and Collibra via Flint Bridge (Enterprise tier). |
Before you enable it
An MCP server runs with whatever access you give it. Start with a development account rather than production, and check which tools the server exposes before you let an agent call them unattended.