Skip to main content Announcing Tool Gateway MCP: the universal MCP Read the announcement
Live 39 Actions

DataGrail MCP Server
for AI Agents

Connect your AI agent to StackOne's DataGrail MCP server and give it 39 MCP tools out of the box. Auth, tool execution, and security all managed.

DataGrail logo
DataGrail MCP Server
Built by StackOne StackOne
DrataGPLocalyzeFlipMindtoolsScreenloop

Coverage

39 Agent Actions

Create, read, update, and delete across DataGrail — and extend your agent's capabilities with custom actions.

Authentication

Agent Tool Authentication

Per-user OAuth in one call. Your DataGrail MCP server gets session-scoped tokens with zero credentials stored on your infra.

Agent Auth →

Security

Agent Protection

Every DataGrail tool response scanned for prompt injection in milliseconds — 88.7% accuracy, all running on CPU.

Prompt Injection Defense →

Performance

Max Agent Context. Min Cost.

Free up to 96% of your agent's context window to enhance reasoning and reduce cost, on every DataGrail call.

Tools Discovery →

What is the DataGrail MCP Server?

A DataGrail MCP server lets AI agents read and write DataGrail data through the Model Context Protocol — Anthropic's open standard for connecting LLMs to external tools. StackOne's DataGrail MCP server ships with 39 pre-built actions, fully extensible via the Connector Builder — plus managed authentication, prompt injection defense, observability, and agent execution runtime. Connect it from MCP clients like Claude Desktop, Claude Code, Cursor, Goose, and VS Code, or from agent frameworks like OpenAI Agents SDK, LangChain, and Vercel AI SDK.

All DataGrail MCP Tools

Every action from DataGrail's API, ready for your agent. Create, read, update, and delete — scoped to exactly what you need.

Assessments

  • Create Assessment

    Create a new assessment from a template, optionally linked to systems. The assessment starts in not_started status. The template must be a Risk Monitor template accessible to the customer. You can optionally attribute the assessment to a creator and invite contributors by email; both fields reference existing active users in your account.

  • List Assessments

    Returns paginated list of assessments with optional filtering. Only returns current versions - superseded assessments are excluded.

  • Get Assessment

    Returns full assessment detail including the form with current answers, per-question visibility flags, progress, and approval details. Use this endpoint to track an assessment through to approval, including who approval is assigned to and who approved it.

Assessment Templates

  • List Assessment Templates

    Returns paginated list of assessment templates available to the customer, including platform defaults.

  • Get Assessment Template

    Returns full assessment template detail including sections and question schema. Divider-type questions are excluded.

Consent Audit Log Exports

  • List Consent Audit Log Exports

    Returns a list of monthly consent audit log export files. Each export covers one calendar month and is updated nightly. Use the export id with the download endpoint to retrieve the file contents.

  • Download Consent Audit Log Export

    Downloads the contents of a specific monthly audit log export file. The response is a JSON file containing an array of individual consent audit log records, returned as an attachment.

Data Discovery Integrations

  • Create Data Discovery Integration

    POST RDD integration details for a system to be scanned using an authenticated agent. If successful, the newly created integration's ID will be returned. Note that names are required to be globally unique. Here, any unique identifier like a hostname, ARN, or service name that is easily associated with a unique system can be used.

  • List Data Discovery Integrations

    GET a list of integrations for an authenticated agent.

  • Get Data Discovery Integration

    GET an RDD integration for an authenticated agent.

  • Update Data Discovery Integration

    UPDATE RDD integration details for a system to be scanned using an authenticated agent. Note that names are required to be globally unique. Here, any unique identifier like a hostname, ARN, or service name that is easily associated with a unique system can be used.

  • Delete Data Discovery Integration

    DELETE an RDD integration for an authenticated agent

Opt Out Integrations

  • List Opt Out Integrations

    Get a list of opt out integrations

  • Update Opt Out Integration

    Update an opt out integration

Opt Out Requests

  • Create Opt Out Request

    Create an opt out request

  • List Opt Out Requests

    Get a list of opt out requests

  • Get Opt Out Request

    Get an opt out request

  • Update Opt Out Request

    Update an opt out request

Privacy Request Integrations

  • List Privacy Request Integrations

    Get a list of privacy request integrations

  • Update Privacy Request Integration

    Update a privacy request integration

Privacy Requests

  • Create Privacy Request

    Create a privacy request

  • List Privacy Requests

    Get a list of privacy requests

  • Get Privacy Request

    Get a privacy request

  • Update Privacy Request

    Update a privacy request

Scan Schedules

  • Create Scan Schedule

    POST a cron expression to perform data discovery scans for a given integration. If successful, the newly created schedule's ID will be returned.

  • List Scan Schedules

    GET a list of schedules for an integration.

  • Get Scan Schedule

    GET a schedule for an integration

  • Update Scan Schedule

    UPDATE the schedule of an integration.

  • Delete Scan Schedule

    DELETE an integration schedule and use DataGrail defaults

Other (10)

  • Upload Assessment Document

    Uploads a document to an assessment. The document is attributed to one of your active users, identified by uploader_email — not to the API key that made the request — so the upload is credited to a person. By default the document is attached to the assessment as a whole; supply tags to attach it to specific questions instead. Send the request as multipart/form-data. The assessment must be in an editable status; uploads to an approved or cancelled assessment are rejected.

  • Upload Files To Integration

    Upload files to the integration

  • Create Detected Objects

    Create Detected Objects

  • Download Assessment PDF

    Download a completed assessment as a PDF document. Only available for assessments in approved status.

  • List Privacy Request Policies

    Get a list of privacy request policies

  • List Security Audit Logs

    Returns login and permission-decision audit log entries for the current customer. With no pagination parameter, up to 1000 records are returned unpaginated; supplying page_size, after, before, or include_total enables pagination (default page size 50). Requires the Login Activity feature.

  • Sync Remote Files To Integration

    Sync remote files to the integration

  • Submit Assessment Answers

    Atomically submits one or more answers for an assessment. All answers are validated before any are persisted — if any answer is invalid, the entire batch is rejected. Requires a last_updated token for optimistic concurrency control.

  • Assign Assessment Approver

    Sets the assessment approver from an email, so a procurement system driving assessments over the API can route sign-off without a user assigning the approver in the app. The email must match an existing active user who is eligible to approve this assessment — one holding an approver role, or a contributor or member of the assessment; otherwise the request is rejected. This only routes approval to the named person — it does not approve the assessment. It cannot be used once the assessment is approved or cancelled.

  • Approve Assessment

    Approves an assessment, so a procurement system driving assessments over the API can sign off without a user opening the app. This performs the same status transition as approving in the app: every required question must be answered first, and the same downstream effects fire. This endpoint is a service-level call, and its authorization differs from approving in the app. In the app, the approval is gated on the signed-in user's own permission to approve. Over the API, the gate is the API token's write access — the token is what is authorized to approve. The approved_by_email you supply is used only for attribution: it records who the approval is credited to, both as the sign-off name and as the actor of the change-history entry (which is marked as API-originated). It can name any active user in your account, including one who would not be permitted to approve the assessment in the app. Any caller holding a write-enabled token can therefore approve on behalf of any active user, so treat the token as the trust boundary. The email must still match exactly one active user in your account, or the request is rejected and nothing is approved. It cannot be used once the assessment is approved or cancelled. Supply the optional last_updated token for optimistic concurrency; omitting it skips the concurrency check and approves the latest assessment state at request time.

Set Up Your DataGrail MCP Server in Minutes

One endpoint. Any framework. Your agent is talking to DataGrail in under 10 lines of code.

Agent Frameworks

Claude Desktop
{
  "mcpServers": {
    "stackone": {
      "command": "npx",
      "args": [
        "-y",
        "mcp-remote@latest",
        "https://api.stackone.com/mcp?x-account-id=<account_id>",
        "--header",
        "Authorization: Basic <YOUR_BASE64_TOKEN>"
      ]
    }
  }
}

DataGrail MCP Server FAQ

Does StackOne have a DataGrail MCP server?
Yes. StackOne offers a hosted DataGrail MCP server with 39 pre-built actions, and every action is tested and QA'd by StackOne. Connect it to Claude, Cursor, and any other MCP client, or to any agent framework through the AI Action SDK. It ships with managed agent authentication, prompt injection defense, and tool discovery with server-side execution that preserve your agent's context window and keep reasoning performance.
DataGrail MCP server vs direct API integration — what's the difference?
A DataGrail MCP server and direct API integration serve different use cases. Direct API integration is for software-to-software — backend code calling DataGrail. A DataGrail MCP server is for AI agents — MCP clients like Claude and Cursor, plus framework agents built with OpenAI, LangChain, or Vercel AI — discovering and calling DataGrail at runtime. StackOne provides both.
How does DataGrail authentication work for AI agents?
DataGrail authentication for AI agents works through a StackOne Connect Session. Create one via the dashboard or the SDK — you get an auth link and ready-to-paste config for Claude Desktop, Cursor, and other MCP clients. Your user authenticates their own DataGrail account; StackOne handles token exchange, storage, and refresh. Credentials never reach the LLM, and each user is isolated via origin_owner_id.
Are DataGrail MCP tools vulnerable to prompt injection?
Yes — DataGrail MCP tools can be vulnerable to indirect prompt injection. Any tool that reads user-written content — documents, messages, tickets, records, or free-text fields — is a potential vector. StackOne Defender scans every tool response before it enters the agent's context — regex patterns in ~1ms, then a MiniLM classifier in ~4ms. 88.7% accuracy, CPU-only.
What is the context bloat of a DataGrail agent and how do I avoid it?
Context bloat happens when DataGrail tool schemas and API responses eat your DataGrail agent's memory, preventing it from reasoning effectively. A single DataGrail query can return a massive JSON response, and connecting multiple tools compounds the problem. Tools Discovery and Code Mode reduce context bloat — loading only relevant tools per query and keeping raw responses out of the agent's context.
Can I limit which actions my DataGrail agent can access?
Yes — you can limit which actions your DataGrail agent can access directly from the StackOne dashboard. Toggle actions on or off, or restrict them to specific accounts, with no code changes to your agent. Session tokens can be scoped to exact actions so if one leaks, exposure stays contained.
Can I create custom agent actions for my DataGrail MCP server?
Yes — you can create custom agent actions for your DataGrail MCP server using Connector Builder. It's an integration agent your coding assistant (Claude Code, Cursor, or Copilot) can invoke to research DataGrail's API, generate production-ready connector YAML, test against the live API, and validate before you ship.
When should I NOT use a DataGrail MCP server?
Skip a DataGrail MCP server if your integration is purely software-to-software — direct DataGrail API integration is simpler when no AI agent is involved. For deterministic, compliance-critical operations (financial transactions, regulatory reporting), direct API gives you predictable behavior without agent-driven decision-making. MCP shines when AI agents need to dynamically discover and call DataGrail actions at runtime.
What AI frameworks and AI clients does the StackOne DataGrail MCP server support?
The StackOne DataGrail MCP server supports both. MCP clients (paste-and-go apps): Claude Desktop, Claude Code, Cursor, VS Code, Goose. Agent frameworks (code SDKs you build with): OpenAI Agents SDK, Anthropic, Vercel AI, Google ADK, CrewAI, Pydantic AI, LangChain, LangGraph, Azure AI Foundry.

Put your AI agents to work

All the tools you need to build and scale AI agent integrations, with best-in-class connectivity, execution, and security.