Claude DataGrail connectorWhat Claude can do in your DataGrail account.
The Claude DataGrail connector exposes 51 tools. 23 read your privacy data, 12 change something in it, 16 are documented nowhere. Below: what Claude looks up, what an admin must unlock before any write, and where the documentation has gaps.
Verified Trustpilot reviews · AI, automation & growth agency
What it changes for a privacy team
Instead of hopping between the request manager, the risk register and the consent console, you ask Claude: how many deletion requests are overdue, which DPIAs are due soon, which cookies still lack a category. It queries DataGrail and answers you right in the chat.
Hit the legal deadlines. search_tickets counts and groups requests by status, type or regulation, and get_ticket_activity_log retraces every step of a disputed case.
Keep the data map current. search_inventory spots detected systems with no assessment, add_system adds one, and associate_processing_activity ties it to the right processing activities.
Sort cookies without publishing by accident. fetch_cookies lists those waiting for a category, then create_cookie_rule sets a draft rule; visitors see nothing until the version is published.
What Claude does not do out of the box: by default, DataGrail opens reads only, and every write tool must be granted person by person by an admin. Sixteen tools on the sheet have no official description at all. And nothing fires when a new request comes in; that kind of monitoring calls for an automation tool, a different job covered on the Integrations hub.
The vocabulary in one minute
Five words worth knowing before you connect DataGrail.
- Connector
- The link you set up once between Claude and an account you already have, so Claude can work in it while it answers you.
- Tool
- One named action a connector opens to Claude. Claude picks which ones to call; the directory lists them by name.
- Authorization
- The service's own sign-in screen, where you hand Claude the access it will use. Granted once per person, and revocable.
- Approval
- The confirmation Claude waits for before an action that changes your account, shown in the chat right when it matters.
- MCP
- The shared standard behind connectors: it is what lets an assistant like Claude talk to an outside service at all.
Connect DataGrail to Claude in three steps
- 01
Find DataGrail in Claude
In Claude's settings, open Customize, then Connectors, and look for DataGrail. On a Team or Enterprise workspace, an Owner or Primary Owner enables it for the organization first, then each person connects.
- 02
Sign in to DataGrail
Click Connect on its row, then sign in through the window DataGrail opens. If the link breaks later, Disconnect and connect again, then re-authorize the tools you need.
- 03
Pick the tools you grant
Read DataGrail's authorization screen carefully: it, not Claude, sets the scope of the access, and it is where you tick the tools to grant. You can add more later by re-authorizing.
The 51 tools, grouped by what they do
DataGrail gives Claude 51 tools: 23 that read your account, 12 that change something in it, and 16 no official source describes.
Three groups: what Claude reads, what it changes, and what no source describes. Names stay exactly as Claude displays them.
- 23 read
- 12 write
- 16 not documented
What Claude reads (23)
23 toolsRequests, consent, risks, inventory, integrations and support, looked up without changing anything.
fetch_integration_errors
Gathers errors from DataGrail integrations, grouped by integration and error type, with sample details to help troubleshoot. You read it straight in the conversation.
fetch_cookies
Shows cookies by state: those waiting for a category, suggested categorization rules, or rules already in place. Claude lays it all out by state.
fetch_tags
Lists tags, meaning the third-party services loaded on your site, that still need a consent category. Claude helps you spot the ones that slipped through.
get_banner_status
Reports the state of consent banners and their containers: publish status, consent categories and the domains they are deployed on. One glance tells you where each site stands.
get_consent_privacy_policies
Returns the policies that govern banner behavior per privacy framework: consent modes, GPC and DNT settings, category configuration. Claude compares the frameworks for you.
get_consent_updates
Traces the consent history over time: publish events, draft revisions and failed publish attempts. You get the sequence without opening the consent console.
get_privacy_request_center
Shows your Privacy Request Center setup: domain, multibrand configuration and, if needed, published intake form revisions with their questions. Useful for rereading what the public sees.
get_request_policies
Provides the privacy request policies configured for your organization: legal frameworks, verification methods and supported rights. Claude answers from the actual settings, not generalities.
search_tickets
Queries privacy request tickets with flexible filters, or counts and groups them through several aggregation modes. Claude can turn the result into a tracking table.
search_opt_outs
Explores Do Not Sell and Do Not Share opt-out requests, with filters and aggregation modes. The answer stays within what your role lets you see.
search_assessment_templates
Browses risk assessment templates, both those DataGrail ships and those your team built. It saves you from starting with a blank page.
search_assessments
Searches completed and in-progress risk assessments (PIAs, DPIAs, vendor reviews), with deadline tracking and contributor filtering. Read access depends on your organization having the matching product enabled and on your role.
search_inventory
Digs through the Live Data Map system inventory, with filters on risk indicators, assessment status and linked processing activities. Results follow your role's permissions.
search_risks
Looks up the register of identified risks and how far their mitigation has gone. Nothing in the register is changed by the lookup.
get_ticket_activity_log
Returns the chronological log of a request ticket: status transitions, emails sent, manual actions. You get a dated trail, step by step.
fetch_privacy_request_integration_stats
Computes privacy request outcomes and trends for one or more integrations, optionally with direct-contact response metrics. Results come back integration by integration, within your permissions.
fetch_available_integrations
Browses DataGrail's full integration catalog, searchable by name, or spots integrations that were discovered but never connected. The answer comes without opening the catalog.
fetch_my_integrations
Details your organization's connected integrations: connection status, capabilities and integration type. It covers only what is connected in your organization.
search_processing_activities
Searches Record of Processing Activities entries, pulls a full record by ID, or finds the activities a given system should be added to. It only reads the register; linking a system is the job of associate_processing_activity.
fetch_data_discovery_agents
Shows your organization's Data Discovery agents, with their connection state and recent scan activity. An admin can still withdraw this read tool from a user.
search_knowledgebase_docs
Runs a semantic search over DataGrail's knowledge base documentation to answer a product question. It searches DataGrail's product documentation, not your organization's privacy data or records.
get_privacy_outcomes_report
Opens the full detail of a Privacy Outcomes Report from its ID, value calculations and operational metrics included. You need the report's ID to open it.
list_support_tickets
Lists the support tickets you yourself filed with DataGrail, so you can follow them from the chat.
What Claude changes (12)
12 toolsAssessments, inventory, cookie rules, tags and support tickets. Each must be granted to you by an admin, and no source gives them a confirmation of their own.
submit_product_feedback
Approval: see the ruleSends product feedback, a feature request or a bug report to DataGrail's product team on your behalf. Per the publisher, it neither modifies nor deletes any existing data.
add_system
Approval: see the ruleAdds a new system to the Live Data Map inventory.
create_assessment
Approval: see the ruleCreates a risk assessment from an existing template.
answer_assessment_question
Approval: see the ruleAnswers a question in an ongoing assessment. Answers are added to the existing assessment.
associate_processing_activity
Approval: see the ruleLinks one or more processing activities to a system in the Live Data Map inventory.
update_assessment
Approval: see the ruleChanges an assessment's name, due date or status.
create_cookie_rule
Approval: see the ruleCreates a draft cookie categorization rule, optionally with a wildcard pattern that covers several cookies.
update_cookie_rule
Approval: see the ruleUpdates an existing cookie categorization rule, still in draft. The rule stays in draft until publication.
delete_cookie_rule
Approval: see the ruleDeletes a draft cookie categorization rule.
update_consent_tag
Approval: see the ruleChanges the category, vendor, description, display name or rights portal of a consent tag on the draft container version.
create_support_ticket
Approval: see the ruleOpens a new support ticket with DataGrail on your behalf.
add_support_ticket_comment
Approval: see the ruleAdds a reply to a support ticket you filed. The reply joins the ticket thread.
Undocumented (16)
16 toolsThe directory publishes these names and nothing else. DataGrail's tool reference does not describe them, and this page will not guess from their names.
fetch_configured_integration_catalog
This name is on the directory sheet, but DataGrail's tool reference, which covers thirty-five other tools, never mentions it. What it does is unknown.
fetch_integration_catalog
Listed by the directory, missing from DataGrail's documentation: no official sentence says what it is for, and this page will not guess from its name. The gap is real, not an oversight on this page.
get_assessment_context
The directory publishes this name and nothing else. Neither the DataGrail tool reference read on October 1, 2026 nor any Claude page covers it.
get_assessment_insights_context
Another name without a description: not the sheet, not DataGrail's reference, not Claude's help center says what it does. That holds across all the documentation read.
save_report_definition
No official source describes this tool. Its name sits on the directory sheet without a single line of explanation from the publisher. No official source fills that gap.
create_opt_out
On the sheet, nowhere in DataGrail's tool reference. Until a source documents it, its effect stays unknown here. No official source describes it to date.
create_direct_contact_integration
DataGrail's documentation never cites it, and the directory sheet offers only its name. Saying more would mean making things up. No official source mentions it.
download_consent_audit_log
The directory announces it, DataGrail's reference ignores it, and no official source documents it. This page therefore attributes no behavior to it at all. Nothing to add while the publisher stays silent.
assign_request
One more name with no official text behind it: neither DataGrail nor Anthropic describes what it does in an account. No official source documents it. Its effect remains unknown.
create_request
It is one of the 16 tools on the sheet that DataGrail's reference leaves out. Its exact purpose is published nowhere. No official source describes it.
add_assessment_contributors
No publisher page documents it. The only public trace is its name in the directory listing. No official source explains its role. Its name is not enough to infer anything from.
add_risk
Only its name is public; no official source explains its role or what it touches inside DataGrail. The gap remains whole. No official source describes it, and the page will say no more.
export_data_classification
Nothing in the documentation read describes this tool, which only shows up on the connector's sheet. No official source documents it, at DataGrail or at Anthropic.
download_system_inventory
Name visible on the sheet, no official source describing it: flagging the gap beats filling it with a guess. Nothing more is published. So the page stops there for this tool.
add_assessment_risk
DataGrail's tool reference says not a word about it. Its existence rests solely on the directory listing. No official source says more. Its role is still for the publisher to document.
add_system_risk
Last of the undocumented tools: the directory lists it, no source describes it, and the page stops there. No official source describes it. Filling that gap is up to the publisher.
What Claude asks before it acts
By default, Claude stops and asks for your go-ahead before each action it takes on an account for you. No source describes a rule specific to any DataGrail tool.
On the Claude side, owners of a Team or Enterprise workspace decide whether a member may let some actions through, and can shut writes for the whole organization. On the DataGrail side, the locks are finer: writes are off by default, an admin grants them tool by tool, and can also withdraw read tools. Every call stays within your role and your organization, and the publisher says each invocation is audit-logged.
Which plans it is available on
None of the 819 sheets in the directory shows plan availability. That information is not published anywhere, connector by connector.
The general rule is published: remote connectors are open to all users on Claude, Cowork, Claude Desktop and mobile. On Team and Enterprise, an Owner or Primary Owner turns the connector on for the organization first. On the DataGrail side, tools also depend on the products your organization has. For the current state on your account, the sheet in the official directory remains the reference.
Where this connector stops
A connector is not an automation. Claude calls these tools while it answers you, and nothing starts on its own when a request lands.
Only the publisher documents this connector: neither Claude's help center nor Anthropic's developer docs have a page on it. The tool list is an observed floor, not a promise, and the partner badge is not a security audit. Anthropic says so on every sheet: it does not choose the tools a publisher exposes and does not guarantee they behave as described. Connect only what comes from a publisher you know.
Need help connecting DataGrail to Claude?
A person reads every message.
