OneTrust Api - A2A | AG-UI | MCP

Version: 1.3.2

Overview

OneTrust Api is a production-grade Python API client, Model Context Protocol (MCP) server, and A2A agent for the OneTrust privacy, consent, data-governance, and risk platform.

It provides 100% coverage of the OneTrust public API — every operation across all 35 OpenAPI specifications (~600 operations, 7 product areas) is exposed as both a typed client method and an action-routed MCP tool. The client, MCP tools, and a machine-readable coverage manifest are all generated from the vendored OpenAPI specs (onetrust_api/specs/*.json) by scripts/generate_from_openapi.py, and a coverage test asserts the three sets stay in lock-step.

Key Features

• 100% Action-Routed MCP Tools — one consolidated tool per domain (e.g. onetrust_incidents, onetrust_dsar, onetrust_assessments) takes an action plus a params_json payload and routes to the underlying API method. 36 tools cover every endpoint without flooding the IDE tool list.
• Full OneTrust surface — AI Governance, Consent & Preference Management, Data Use Governance, Privacy Automation (DSAR, Assessments, Data Mapping, Incidents), Tech Risk & Compliance, Third-Party Management, ESG, and Platform.
• Flexible auth — a pre-minted OAuth2 bearer token or the OAuth2 client-credentials flow (auto-exchanged and refreshed), plus OIDC delegation (RFC 8693) via agent-connector-sdk.
• Multi-region / multi-service aware — regional tenant pods, the consent privacy-portal host, and on-prem worker nodes are resolved per-operation.
• Resilient — honours 429 Retry-After, retries transient 5xx, and handles both OneTrust pagination styles (offset and cursor).

MCP

Using as an MCP Server

The MCP Server runs in stdio (local) or streamable-http (networked) mode. Each domain is a tool gated by a {TAG}TOOL environment variable (default True), so the operator can scope the surface (e.g. set ESGTOOL=False to drop ESG).

Environment Variables

Package environment variables

Variable	Example	Description
ONETRUST_URL	—	─── OneTrust connection ─────────────────────────────────────────────── Tenant host URL (overrides ONETRUST_REGION). e.g. https://acme.my.onetrust.com
ONETRUST_REGION	us	Or pick a shared regional pod: us, eu, de, uk, au, ca, fr, in, jp, trial, uat ...
ONETRUST_TOKEN	secret-injected	─── Authentication (use ONE of the two) ─────────────────────────────── 1) Pre-minted OAuth2 bearer token (Global Settings > Access Management > Credentials)
ONETRUST_CLIENT_ID	—	2) OAuth2 client-credentials (exchanged at /api/access/v1/oauth/token)
ONETRUST_CLIENT_SECRET	secret-injected
ONETRUST_CONSENT_URL	—	─── Optional service hosts ──────────────────────────────────────────── Consent transaction / privacy-portal host (consent_receipts, universal_consent ...)
ONETRUST_WORKER_URL	—	On-prem Data Discovery worker-node host
TLS_PROFILE	private-pki	─── HTTP behaviour ──────────────────────────────────────────────────── TLS verification is mandatory. Select a named runtime profile from AgentConfig.
TLS_PROFILES_REF	secret://runtime/tls-profiles
FASTMCP_LOG_LEVEL	INFO	─── MCP transport / auth (agent-connector-sdk) ────────────────────────────
TRANSPORT	stdio
AUTH_TYPE	none
ACCESS_MANAGEMENTTOOL	True	─── Tool Toggle Switches (per-domain
AI_GOVERNANCETOOL	True
CUSTOM_APITOOL	True
ASSESSMENTSTOOL	True
AUDIT_MANAGEMENTTOOL	True
BULK_EXPORTTOOL	True
CMPTOOL	True
COMPLIANCE_AUTOMATIONTOOL	True
CONSENT_INTERFACESTOOL	True
CONSENT_RECEIPTSTOOL	True
COOKIE_CONSENTTOOL	True
COOKIE_CONSENT_LEGACYTOOL	True
COOKIE_DOMAIN_DATATOOL	True
CROSS_DEVICE_CONSENTTOOL	True
DATA_CATALOGTOOL	True
DATA_DISCOVERYTOOL	True
DATA_DISCOVERY_WORKERTOOL	True
DATA_MAPPINGTOOL	True
DATA_MAPPING_LEGACYTOOL	True
DOCUMENTSTOOL	True
DSARTOOL	True
ESGTOOL	True
INCIDENTSTOOL	True
INTEGRATIONSTOOL	True
INVENTORYTOOL	True
ISSUES_MANAGEMENTTOOL	True
IT_RISK_MANAGEMENTTOOL	True
MOBILE_APP_CONSENTTOOL	True
OBJECT_MANAGERTOOL	True
POLICY_MANAGEMENTTOOL	True
PRIVACY_NOTICESTOOL	True
TASK_MANAGEMENTTOOL	True
TPRMTOOL	True
TRAININGTOOL	True
UNIVERSAL_CONSENTTOOL	True
USER_PROVISIONINGTOOL	True
ONETRUST_API_MCP_IMAGE	—	─── Deployment images ──────────────────────────────────────────────── Supply operator-approved immutable image@sha256 references for deployment.
ONETRUST_API_AGENT_IMAGE	—
ONETRUST_TLS_PROFILE	—	─── Service-specific TLS selection ──────────────────────────────────── Optional named profile and runtime profile reference; verification stays mandatory.
ONETRUST_TLS_PROFILE_REF	—
INGESTTOOL	—	The drift scanner derives this name from register_ingest_tools, but this connector registers ingest tools directly and does not read this toggle. This is documentation of that limitation, not a working disable switch.

Inherited agent-connector-sdk variables (apply to every connector)

Variable	Example	Description
HOST	127.0.0.1	Loopback bind host (set an authenticated ingress explicitly)
PORT	8000	Bind port (HTTP transports)
MCP_ENABLED_TOOLS	—	Comma-separated tool allow-list
MCP_DISABLED_TOOLS	—	Comma-separated tool deny-list
MCP_ENABLED_TAGS	—	Comma-separated tag allow-list
MCP_DISABLED_TAGS	—	Comma-separated tag deny-list
EUNOMIA_TYPE	none	Authorization mode: none | embedded | remote
EUNOMIA_POLICY_FILE	mcp_policies.json	Embedded Eunomia policy file
EUNOMIA_REMOTE_URL	—	Remote Eunomia authorization server URL
ENABLE_OTEL	False	Enable OpenTelemetry export
OTEL_EXPORTER_OTLP_ENDPOINT	—	OTLP collector endpoint
MCP_CLIENT_AUTH	—	Outbound MCP child auth: oidc-client-credentials | basic | none
OIDC_CLIENT_ID	—	OIDC client id (service-account auth)
OIDC_CLIENT_SECRET_REF	secret://identity/oidc-client-secret	Runtime secret reference for the OIDC service account
MCP_BASIC_AUTH_USERNAME	—	HTTP Basic username (MCP_CLIENT_AUTH=basic)
MCP_BASIC_AUTH_PASSWORD_REF	secret://identity/mcp-basic-password	Runtime secret reference for HTTP Basic auth (MCP_CLIENT_AUTH=basic)
DEBUG	False	Verbose logging
PYTHONUNBUFFERED	1	Unbuffered stdout (recommended in containers)
MCP_URL	http://localhost:8000/mcp	URL of the MCP server the agent connects to

53 package + 22 inherited variable(s). Auto-generated from .env.example + the shared agent-connector-sdk set — do not edit.

Variable	Description
ONETRUST_URL	Tenant host URL, e.g. https://acme.my.onetrust.com (overrides region).
ONETRUST_REGION	Shared pod when no URL is set: us, eu, de, uk, au, ca, fr, in, jp, trial, uat, … (default us).
ONETRUST_TOKEN	Pre-minted OAuth2 bearer token.
ONETRUST_CLIENT_ID / ONETRUST_CLIENT_SECRET	OAuth2 client-credentials (exchanged at /api/access/v1/oauth/token).
ONETRUST_CONSENT_URL	Optional host for consent-transaction APIs (privacy portal).
ONETRUST_WORKER_URL	Optional on-prem Data Discovery worker-node host.
TLS_PROFILE	Named AgentConfig transport-security profile; verification is mandatory.
TLS_PROFILES_REF	Runtime secret reference for the TLS profile catalog.
<DOMAIN>TOOL	Toggle a domain tool, e.g. INCIDENTSTOOL, DSARTOOL, CONSENT_RECEIPTSTOOL (default True).

Run in stdio mode (default):

export ONETRUST_URL="https://acme.my.onetrust.com"
export ONETRUST_TOKEN="your_token"
onetrust-mcp --transport "stdio"

Run in HTTP mode:

export ONETRUST_URL="https://acme.my.onetrust.com"
export ONETRUST_TOKEN="your_token"
onetrust-mcp --transport "streamable-http" --host "0.0.0.0" --port "8000"

Tool Domains

access_management, ai_governance, assessments, audit_management, bulk_export, cmp, compliance_automation, consent_interfaces, consent_receipts, cookie_consent, cookie_consent_legacy, cookie_domain_data, cross_device_consent, data_catalog, data_discovery, data_discovery_worker, data_mapping, data_mapping_legacy, documents, dsar, esg, incidents, integrations, inventory, issues_management, it_risk_management, mobile_app_consent, object_manager, policy_management, privacy_notices, task_management, tprm, training, universal_consent, user_provisioning — plus custom_api (a raw REST escape hatch).

Available MCP Tools

Intent tools

MCP Tool	Kind	Description
onetrust_access_management	intent	Manage OneTrust access management operations.
onetrust_ai_governance	intent	Manage OneTrust ai governance operations.
onetrust_api_request	intent	Execute an arbitrary OneTrust REST API request directly.
onetrust_assessments	intent	Manage OneTrust assessments operations.
onetrust_audit_management	intent	Manage OneTrust audit management operations.
onetrust_bulk_export	intent	Manage OneTrust bulk export operations.
onetrust_cmp	intent	Manage OneTrust cmp operations.
onetrust_compliance_automation	intent	Manage OneTrust compliance automation operations.
onetrust_consent_interfaces	intent	Manage OneTrust consent interfaces operations.
onetrust_consent_receipts	intent	Manage OneTrust consent receipts operations.
onetrust_cookie_consent	intent	Manage OneTrust cookie consent operations.
onetrust_cookie_consent_legacy	intent	Manage OneTrust cookie consent legacy operations.
onetrust_cookie_domain_data	intent	Manage OneTrust cookie domain data operations.
onetrust_cross_device_consent	intent	Manage OneTrust cross device consent operations.
onetrust_data_catalog	intent	Manage OneTrust data catalog operations.
onetrust_data_discovery	intent	Manage OneTrust data discovery operations.
onetrust_data_discovery_worker	intent	Manage OneTrust data discovery worker operations.
onetrust_data_mapping	intent	Manage OneTrust data mapping operations.
onetrust_data_mapping_legacy	intent	Manage OneTrust data mapping legacy operations.
onetrust_documents	intent	Manage OneTrust documents operations.
onetrust_dsar	intent	Manage OneTrust dsar operations.
onetrust_esg	intent	Manage OneTrust esg operations.
onetrust_incidents	intent	Manage OneTrust incidents operations.
onetrust_ingest_assessments	intent	Ingest OneTrust assessments into the KG as typed :Assessment nodes + :Document summaries.
onetrust_ingest_cookies	intent	Ingest OneTrust cookie-scan results into the KG as typed :Cookie (+ :CookieDomain) nodes.
onetrust_ingest_inventories	intent	Ingest OneTrust data-inventory records into the KG as typed :Inventory (+ :DataElement) nodes.
onetrust_integrations	intent	Manage OneTrust integrations operations.
onetrust_inventory	intent	Manage OneTrust inventory operations.
onetrust_issues_management	intent	Manage OneTrust issues management operations.
onetrust_it_risk_management	intent	Manage OneTrust it risk management operations.
onetrust_mobile_app_consent	intent	Manage OneTrust mobile app consent operations.
onetrust_object_manager	intent	Manage OneTrust object manager operations.
onetrust_policy_management	intent	Manage OneTrust policy management operations.
onetrust_privacy_notices	intent	Manage OneTrust privacy notices operations.
onetrust_task_management	intent	Manage OneTrust task management operations.
onetrust_tprm	intent	Manage OneTrust tprm operations.
onetrust_training	intent	Manage OneTrust training operations.
onetrust_universal_consent	intent	Manage OneTrust universal consent operations.
onetrust_user_provisioning	intent	Manage OneTrust user provisioning operations.

39 intent tool(s) · 0 operation(s). Call an operation as act(action='<operation>', params={...}) (ask for reads); find(action='describe', params={'action': '<operation>'}) returns its arguments. An operation is enabled unless its <DOMAIN>TOOL toggle is set false. Auto-generated — do not edit.

MCP Configuration Examples

Install the connector-focused [mcp] extra. Examples use onetrust-api[mcp] to add FastMCP / FastAPI through agent-connector-sdk[mcp]; the required Agent Utilities core still carries epistemic-graph[full]. The [agent-runtime] extra additionally enables model orchestration.

stdio Transport (local IDEs — Cursor, Claude Desktop, VS Code)

{
  "mcpServers": {
    "onetrust-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "onetrust-api[mcp]",
        "onetrust-mcp"
      ],
      "env": {
        "ACCESS_MANAGEMENTTOOL": "True",
        "AI_GOVERNANCETOOL": "True",
        "ASSESSMENTSTOOL": "True",
        "AUDIT_MANAGEMENTTOOL": "True",
        "BULK_EXPORTTOOL": "True",
        "CMPTOOL": "True",
        "COMPLIANCE_AUTOMATIONTOOL": "True",
        "CONSENT_INTERFACESTOOL": "True",
        "CONSENT_RECEIPTSTOOL": "True",
        "COOKIE_CONSENTTOOL": "True",
        "COOKIE_CONSENT_LEGACYTOOL": "True",
        "COOKIE_DOMAIN_DATATOOL": "True",
        "CROSS_DEVICE_CONSENTTOOL": "True",
        "CUSTOM_APITOOL": "True",
        "DATA_CATALOGTOOL": "True",
        "DATA_DISCOVERYTOOL": "True",
        "DATA_DISCOVERY_WORKERTOOL": "True",
        "DATA_MAPPINGTOOL": "True",
        "DATA_MAPPING_LEGACYTOOL": "True",
        "DOCUMENTSTOOL": "True",
        "DSARTOOL": "True",
        "ESGTOOL": "True",
        "INCIDENTSTOOL": "True",
        "INTEGRATIONSTOOL": "True",
        "INVENTORYTOOL": "True",
        "ISSUES_MANAGEMENTTOOL": "True",
        "IT_RISK_MANAGEMENTTOOL": "True",
        "MOBILE_APP_CONSENTTOOL": "True",
        "OBJECT_MANAGERTOOL": "True",
        "ONETRUST_REGION": "us",
        "POLICY_MANAGEMENTTOOL": "True",
        "PRIVACY_NOTICESTOOL": "True",
        "TASK_MANAGEMENTTOOL": "True",
        "TPRMTOOL": "True",
        "TRAININGTOOL": "True",
        "UNIVERSAL_CONSENTTOOL": "True",
        "USER_PROVISIONINGTOOL": "True"
      }
    }
  }
}

Runtime references require an alias-aware launcher such as Arbora. Other launchers must omit those entries and inject the resolved values through their own runtime secret boundary.

Streamable-HTTP Transport (networked / production)

{
  "mcpServers": {
    "onetrust-mcp": {
      "command": "uvx",
      "args": [
        "--from",
        "onetrust-api[mcp]",
        "onetrust-mcp",
        "--transport",
        "streamable-http",
        "--port",
        "8000"
      ],
      "env": {
        "TRANSPORT": "streamable-http",
        "HOST": "127.0.0.1",
        "PORT": "8000",
        "ACCESS_MANAGEMENTTOOL": "True",
        "AI_GOVERNANCETOOL": "True",
        "ASSESSMENTSTOOL": "True",
        "AUDIT_MANAGEMENTTOOL": "True",
        "BULK_EXPORTTOOL": "True",
        "CMPTOOL": "True",
        "COMPLIANCE_AUTOMATIONTOOL": "True",
        "CONSENT_INTERFACESTOOL": "True",
        "CONSENT_RECEIPTSTOOL": "True",
        "COOKIE_CONSENTTOOL": "True",
        "COOKIE_CONSENT_LEGACYTOOL": "True",
        "COOKIE_DOMAIN_DATATOOL": "True",
        "CROSS_DEVICE_CONSENTTOOL": "True",
        "CUSTOM_APITOOL": "True",
        "DATA_CATALOGTOOL": "True",
        "DATA_DISCOVERYTOOL": "True",
        "DATA_DISCOVERY_WORKERTOOL": "True",
        "DATA_MAPPINGTOOL": "True",
        "DATA_MAPPING_LEGACYTOOL": "True",
        "DOCUMENTSTOOL": "True",
        "DSARTOOL": "True",
        "ESGTOOL": "True",
        "INCIDENTSTOOL": "True",
        "INTEGRATIONSTOOL": "True",
        "INVENTORYTOOL": "True",
        "ISSUES_MANAGEMENTTOOL": "True",
        "IT_RISK_MANAGEMENTTOOL": "True",
        "MOBILE_APP_CONSENTTOOL": "True",
        "OBJECT_MANAGERTOOL": "True",
        "ONETRUST_REGION": "us",
        "POLICY_MANAGEMENTTOOL": "True",
        "PRIVACY_NOTICESTOOL": "True",
        "TASK_MANAGEMENTTOOL": "True",
        "TPRMTOOL": "True",
        "TRAININGTOOL": "True",
        "UNIVERSAL_CONSENTTOOL": "True",
        "USER_PROVISIONINGTOOL": "True"
      }
    }
  }
}

Alternatively, connect to a pre-deployed Streamable-HTTP instance by url:

{
  "mcpServers": {
    "onetrust-mcp": {
      "url": "http://localhost:8000/onetrust-mcp/mcp"
    }
  }
}

Run a reviewed container image as a least-privilege stdio child (no listener or published port):

docker run -i --rm \
  --read-only \
  --cap-drop=ALL \
  --security-opt=no-new-privileges \
  --pids-limit=256 \
  --tmpfs /tmp:rw,noexec,nosuid,nodev,size=64m \
  -e TRANSPORT=stdio \
  -e ACCESS_MANAGEMENTTOOL=True \
  -e AI_GOVERNANCETOOL=True \
  -e ASSESSMENTSTOOL=True \
  -e AUDIT_MANAGEMENTTOOL=True \
  -e BULK_EXPORTTOOL=True \
  -e CMPTOOL=True \
  -e COMPLIANCE_AUTOMATIONTOOL=True \
  -e CONSENT_INTERFACESTOOL=True \
  -e CONSENT_RECEIPTSTOOL=True \
  -e COOKIE_CONSENTTOOL=True \
  -e COOKIE_CONSENT_LEGACYTOOL=True \
  -e COOKIE_DOMAIN_DATATOOL=True \
  -e CROSS_DEVICE_CONSENTTOOL=True \
  -e CUSTOM_APITOOL=True \
  -e DATA_CATALOGTOOL=True \
  -e DATA_DISCOVERYTOOL=True \
  -e DATA_DISCOVERY_WORKERTOOL=True \
  -e DATA_MAPPINGTOOL=True \
  -e DATA_MAPPING_LEGACYTOOL=True \
  -e DOCUMENTSTOOL=True \
  -e DSARTOOL=True \
  -e ESGTOOL=True \
  -e INCIDENTSTOOL=True \
  -e INTEGRATIONSTOOL=True \
  -e INVENTORYTOOL=True \
  -e ISSUES_MANAGEMENTTOOL=True \
  -e IT_RISK_MANAGEMENTTOOL=True \
  -e MOBILE_APP_CONSENTTOOL=True \
  -e OBJECT_MANAGERTOOL=True \
  -e ONETRUST_REGION=us \
  -e POLICY_MANAGEMENTTOOL=True \
  -e PRIVACY_NOTICESTOOL=True \
  -e TASK_MANAGEMENTTOOL=True \
  -e TPRMTOOL=True \
  -e TRAININGTOOL=True \
  -e UNIVERSAL_CONSENTTOOL=True \
  -e USER_PROVISIONINGTOOL=True \
  registry.example.invalid/onetrust-api@sha256:<digest> onetrust-mcp

For containerized network HTTP, supply an authenticated TLS ingress (or direct server TLS), exact MCP_ALLOWED_HOSTS, and an exact trusted-proxy CIDR policy through the operator-owned deployment profile. The generator does not emit an unauthenticated non-loopback listener.

Auto-generated from the code-read env surface (package vars) — do not edit.

A2A Agent

Run A2A Server

export ONETRUST_URL="https://acme.my.onetrust.com"
export ONETRUST_TOKEN="your_token"
onetrust-agent --provider openai --model-id gpt-4o --api-key sk-...

Docker

Build

docker build -t onetrust-api .

Run MCP Server

docker run -d \
  --name onetrust-api-mcp \
  -p 8000:8000 \
  -e TRANSPORT=streamable-http \
  -e PORT=8000 \
  -e ONETRUST_URL="https://acme.my.onetrust.com" \
  -e ONETRUST_TOKEN="your_token" \
  example/onetrust-api:mcp

The :mcp tag is the MCP-serving image (built from docker/Dockerfile --target mcp, installing onetrust-api[mcp]). The default the immutable agent image is the full agent image (--target agent, onetrust-api[agent]) which also bundles the Pydantic AI agent and the epistemic-graph engine — use it when the operator run onetrust-agent (the agent), not just the MCP server. See Container images.

Deploy with Docker Compose

services:
  onetrust-api-mcp:
    image: example/onetrust-api:mcp
    environment:
      - HOST=0.0.0.0
      - PORT=8000
      - TRANSPORT=streamable-http
      - ONETRUST_URL=https://acme.my.onetrust.com
      - ONETRUST_TOKEN=your_token
    ports:
      - 8000:8000

Installation

Pick the extra that matches what the operator want to run:

Extra	Installs	Use when
onetrust-api[mcp]	Connector-focused MCP server (agent-connector-sdk[mcp] — FastMCP/FastAPI + epistemic-graph[full])	The operator only run the MCP server (smallest install / image)
onetrust-api[agent]	Agent runtime (agent-connector-sdk[agent-runtime,logfire] — model orchestration + epistemic-graph[full])	The operator run the integrated agent
onetrust-api[all]	Everything (mcp + agent)	Development / both surfaces

# Connector-focused MCP server (includes the shared graph engine)
uv pip install "onetrust-api[mcp]"

# Agent runtime (adds model orchestration to the shared graph engine)
uv pip install "onetrust-api[agent]"

# Everything (development)
uv pip install "onetrust-api[all]"      # or: python -m pip install "onetrust-api[all]"

Container images (:mcp vs :agent)

One multi-stage docker/Dockerfile builds two right-sized images, selected by --target:

Image tag	Build target	Contents	Entrypoint
example/onetrust-api:mcp	--target mcp	onetrust-api[mcp] — connector-focused, includes epistemic-graph[full]; no model-orchestration stack	onetrust-mcp
example/onetrust-api@sha256:<digest>	--target agent (default)	onetrust-api[agent] — agent runtime, model orchestration + epistemic-graph[full]	onetrust-agent

docker build --target mcp   -t example/onetrust-api:mcp    docker/   # connector-focused MCP server
docker build --target agent -t example/onetrust-api:agent-local docker/   # agent runtime

docker/mcp.compose.yml runs the connector-focused :mcp server; docker/agent.compose.yml runs the agent (immutable agent digest) with a co-located :mcp sidecar.

Knowledge-graph database (epistemic-graph)

Both [mcp] and [agent] carry the epistemic-graph engine through the required Agent Utilities core dependency (epistemic-graph[full]). The [mcp] extra keeps the server connector-focused; [agent] also enables model orchestration. Local deployments can use the bundled engine. For production or shared state, run epistemic-graph as a dedicated database service and configure the runtime to use it. Deployment recipes (single-node + Raft HA), connection configuration, and architecture diagrams are documented in the epistemic-graph deployment guide.

Environment Variables

Every variable the server reads. See .env.example for a copy-paste starting point.

Connection & credentials (OneTrust)

Variable	Description	Default
ONETRUST_URL	Tenant host URL, e.g. https://acme.my.onetrust.com (overrides ONETRUST_REGION)	—
ONETRUST_REGION	Shared regional pod when no URL is set: us, eu, de, uk, au, ca, fr, in, jp, trial, uat, …	us
ONETRUST_TOKEN	Pre-minted OAuth2 bearer token	—
ONETRUST_CLIENT_ID	OAuth2 client-credentials client id (exchanged at /api/access/v1/oauth/token)	—
ONETRUST_CLIENT_SECRET	OAuth2 client-credentials client secret	—
ONETRUST_CONSENT_URL	Optional host for consent-transaction APIs (privacy portal)	—
ONETRUST_WORKER_URL	Optional on-prem Data Discovery worker-node host	—
TLS_PROFILE	Named AgentConfig transport-security profile; verification is mandatory	—
TLS_PROFILES_REF	Runtime secret reference for the TLS profile catalog	—

MCP server / transport

Variable	Description	Default
TRANSPORT	stdio, streamable-http, or sse	stdio
HOST	Bind host (HTTP transports)	0.0.0.0
PORT	Bind port (HTTP transports)	8000
AUTH_TYPE	MCP transport auth mode (none, …)	none
MCP_TOOL_MODE	Tool surface: condensed, verbose, or both	condensed
MCP_ENABLED_TOOLS / MCP_DISABLED_TOOLS	Comma-separated tool allow/deny list	—
MCP_ENABLED_TAGS / MCP_DISABLED_TAGS	Comma-separated tag allow/deny list	—
FASTMCP_LOG_LEVEL	FastMCP log level (e.g. INFO, DEBUG)	INFO
DEBUG	Verbose logging	False
PYTHONUNBUFFERED	Unbuffered stdout (recommended in containers)	1

Tool toggles

Each action-routed domain tool can be disabled individually via its <DOMAIN>TOOL toggle env var (set to false). The full list is in the Available MCP Tools table above (e.g. INCIDENTSTOOL, DSARTOOL, CONSENT_RECEIPTSTOOL, ESGTOOL).

Telemetry & governance

Variable	Description	Default
ENABLE_OTEL	Enable OpenTelemetry export	True
OTEL_EXPORTER_OTLP_ENDPOINT	OTLP collector endpoint	—
OTEL_EXPORTER_OTLP_PUBLIC_KEY / OTEL_EXPORTER_OTLP_SECRET_KEY	OTLP auth keys	—
OTEL_EXPORTER_OTLP_PROTOCOL	OTLP protocol (e.g. http/protobuf)	—
EUNOMIA_TYPE	Authorization mode: none, embedded, remote	none
EUNOMIA_POLICY_FILE	Embedded policy file	mcp_policies.json
EUNOMIA_REMOTE_URL	Remote Eunomia server URL	—

Agent CLI (full [agent] runtime only)

Variable	Description	Default
MCP_URL	URL of the MCP server the agent connects to	http://localhost:8000/mcp

Documentation

The complete documentation is published as the official documentation site and is the source of truth for installation, usage, and deployment.

Page	Covers
Overview	the action-routed tool surface and architecture
Installation	pip, source, extras, prebuilt Docker image
Usage (API / CLI / MCP)	the MCP tools, the Api client, the CLI
Deployment	run the MCP and agent servers, Compose, env config

Repository Owners

Deploy with agent-connector-sdk-deployment

Provision this package with the consolidated agent-connector-sdk-deployment workflow. It selects an installed-package, editable-source, or immutable-container path; records only runtime secret and TLS-profile references in AgentConfig; and runs doctor, registration, policy, observability, and rollback gates. Ask your agent to "deploy onetrust-api with agent-connector-sdk-deployment".

Install mode	Command
Installed package	uv tool install "onetrust-api[mcp]", then run onetrust-mcp
Editable source	uv pip install -e ".[agent]", then run onetrust-mcp
Immutable container	deploy registry.example.invalid/onetrust-api@sha256:<digest> through the operator-selected orchestrator

The repository embeds no deployment profile, credential value, certificate path, or environment-specific endpoint. Supply those at runtime through AgentConfig and the configured secret provider.

Governed capability contract

This package ships a compact canonical skill surface with specialist procedures kept as referenced workflows. The current MCP tools, skill metadata, connector_manifest.yml, ontology, mappings, shapes, fixtures, migrations, tool-schema fingerprints, and certification metadata form one versioned capability contract. Validate them together; do not rely on stale tool names or historical per-task skill wrappers.

Runtime endpoints, credentials, certificate trust, tenant identity, retention, and observability policy are deployment inputs and are never packaged values. See Configuration, trust, and privacy before enabling a network transport, connector ingestion, Arbora delegation, or trace export.