Resources · Claude connector

Claude Enterpret connectorWhat Claude can do in your Enterpret account.

The Claude Enterpret connector exposes 7 tools. All 7 read your customer feedback, 0 write, 0 are left undescribed. Below: what Claude can ask your feedback graph, how its answers stay linked to real records, and who controls access.

Verified Trustpilot reviews · AI, automation & growth agency

Overview

What changes when Claude can query your feedback

Enterpret pulls support tickets, Gong calls, app reviews, NPS surveys and Slack threads into one knowledge graph. With the connector, you stop building dashboard filters to answer a product question. You ask Claude in plain words, it runs the query itself, and each number it gives links back to the records behind it in your Enterpret dashboard.

Size a problem with evidence attached. How many enterprise accounts complained about onboarding this quarter? run_graph_query pulls the matching feedback, and every count in the answer is a link you can open to check.

Quote customers in their own words. Before a roadmap review, you want what users actually said about exports. find_user_quote returns verbatim quotes taken from the records behind the answer. If no validated quote exists, Enterpret says the assistant tells you so rather than writing one.

Filter on the labels your team really uses. search_graph_values finds the exact theme, account or source names, and search_graph_fields surfaces source-specific data such as Zendesk tags or App Store ratings, so the query does not miss because of a spelling.

What it will not do: tag feedback, create themes or change anything in Enterpret. By default the sheet only exposes reading. Nothing runs on its own either; Claude calls these tools while it answers you. For a connector that writes to a work tool, see the Claude asana connector page.

Vocabulary

The vocabulary in one minute

Five words you will meet while connecting Enterpret.

Connector
The link you set up once between Claude and an account you already own, so Claude can work in it while it answers you.
Tool
One named thing a connector lets Claude do. Claude picks the ones it needs on its own; the directory sheet lists them by name.
Authorization
Enterpret's own sign-in screen, where you hand Claude the access it will use afterwards. Granted once per person, and revocable.
Directory
The official catalogue where every connector has a sheet with its publisher, its purpose and the tools it exposes.
MCP
The shared standard connectors are built on, which is what lets an assistant like Claude talk to an outside service such as Enterpret.
Connect

Connect Enterpret to Claude in three steps

  1. 01

    Find Enterpret in the connector list

    In Claude's settings, open Customize, then Connectors, and look for Enterpret. On a Team or Enterprise workspace, an Owner or Primary Owner enables it for the organization first; then each member connects on their own.

  2. 02

    Launch the connection

    Click Connect on its row and sign in to Enterpret in the window the service opens. Each person authenticates with their own account, so queries are tracked per member.

  3. 03

    Read the authorization screen

    Check Enterpret's screen before you approve. It belongs to the service, not to Claude, and it decides what the access covers. If the link breaks, Disconnect and connect again; access can also be withdrawn from the Enterpret side.

Tools

The 7 tools, sorted by what they do

Enterpret gives Claude 7 tools: 7 that read your account, 0 that change something in it.

A single group this time: all seven tools read, none changes your data. Names stay exactly as Claude shows them, and you never need to call them yourself.

  • 7 read

What Claude reads (7)

7 tools

Seven tools that look up, filter and quote your feedback without modifying it.

get_organization_details

Identifies which Enterpret organization you are connected to, so the links in Claude's answers open the right dashboard.

When it helps
you work with several Enterpret organizations and want to be sure the numbers come from the correct one.
Watch out
each organization needs its own connection; one link does not span them all.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

get_graph_schema

Shows Claude what kinds of feedback you track and how they are organized, before it builds a question on top of them.

When it helps
a new PM asks what data is even in Enterpret before planning research.
Watch out
this is a map of your data, not the data itself; the actual records come from a query.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

run_graph_query

Pulls the feedback that answers your question and attaches citation links back to the source records in Enterpret.

When it helps
leadership wants the share of churned accounts that mentioned pricing, with proof they can click.
Watch out
a broad question gives a broad answer; naming a theme or a source sharpens it.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

search_graph_values

Finds the precise names of themes, accounts, sources or taxonomy values in your data, so Claude filters on real labels instead of guesses.

When it helps
your team calls it "SSO issues" but the taxonomy says "Single sign-on", and the query must match.
Watch out
it returns labels, not feedback; the query step comes after.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

get_query_examples

Gives Claude reference patterns for common kinds of feedback questions before it builds its own.

When it helps
you ask something unusual, like a week-by-week trend on one feature, and want the query shaped properly.
Watch out
Claude uses these on its own; you will see the name in the chat without having asked for it.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

search_graph_fields

Surfaces data points that belong to one integration, such as Zendesk tags, App Store ratings or Gong call metadata.

When it helps
you want only one-star App Store reviews, or Gong calls from a given sales stage.
Watch out
a field only exists if that source is connected to your Enterpret workspace.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

find_user_quote

Returns verbatim customer quotes taken from the records behind Claude's answer.

When it helps
a product brief needs three real sentences from users about the mobile app, not a paraphrase.
Watch out
Enterpret says that without a validated quote, the assistant tells you none exists rather than inventing one.

Sourcehelpcenter.enterpret.com · May 31, 2026 ↗

Approvals

What Claude asks before it acts

By default, Claude stops and asks before any action it takes on your behalf. With Enterpret, the documented tools only read, so there is no change to confirm.

On a Team or Enterprise workspace, owners decide whether members can let some actions through without being asked each time, and they can limit what a connector may do for the whole organization. That setting applies to everyone and nobody overrides it from their own account. Claude also works with your Enterpret access and nothing more: each member signs in with their own account, and usage is tracked person by person.

Plans

Which plans include it

None of the 819 sheets in the official directory shows availability by plan. That answer, connector by connector, is published nowhere; it is a real gap in the catalogue.

Anthropic's 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 enables a connector for the organization before members connect. Enterpret's own article says something stricter, set against this rule in the limits section. The Enterpret sheet in the official directory shows the current status.

Limits

Where this connector stops

A connector is not an automation. Claude calls these tools while it answers you; no digest goes out on its own when a new wave of tickets arrives.

By default, the sheet only exposes reading, which is an observed floor rather than a guarantee: an administrator can open actions no public sheet shows. The partner badge is not a security audit, and Anthropic says on every sheet that it neither picks a publisher's tools nor guarantees how they behave. This connector is documented by Enterpret alone. Other connector pages sit in the Integrations hub.

Two official sources disagree

Which Claude plans can use the Enterpret connector?

What this page followsEnterpret is talking about connectors in general, not only its own, and its page (May 2026) predates Anthropic's rule. Neither settles it for your account: check the Enterpret sheet in the directory on your own account.

Need help

Need help connecting Enterpret to Claude?

A person reads every message.

FAQ

Frequently asked questions about the Claude Enterpret connector

01What can Claude do with the Enterpret connector?
Claude can question your unified customer feedback and answer with evidence. Its seven tools identify your organization, map how your data is organized, suggest query patterns, find exact labels and source-specific fields, run the query itself, and pull verbatim quotes. Every aggregate number in an answer links back to the matching records in your Enterpret dashboard, so you can check what drove it. You ask in plain language; there is no query syntax to learn.
02Can Claude change or delete anything in Enterpret?
Not with the tools published today. All seven tools on the directory sheet read, and Enterpret's own list describes none that tags feedback, creates a theme or deletes a record. That is a floor observed on the public sheet rather than a promise: an administrator can open actions that no public page shows. If you need to change the taxonomy or the data, that still happens in Enterpret itself. For now, think of it as a research assistant, not an editor.
03Does Claude ask before running queries on my feedback?
By default, Claude asks for confirmation before any action it takes on your behalf, but these tools only read, so there is no change to confirm. On a Team or Enterprise workspace, owners decide whether members may let some actions through without a prompt, and they can restrict the connector for the whole organization. Each member also signs in separately, so what Claude runs is tracked under the person who asked.
04Which Claude plans include the Enterpret connector?
No official source publishes plan availability connector by connector, and none of the 819 directory sheets shows it. Anthropic's published rule says remote connectors are open to all users, and that on Team and Enterprise an Owner or Primary Owner enables a connector first. Enterpret's help article states a stricter condition, which this page sets against Anthropic's rule. The Enterpret sheet in the directory is the place to check your status.
05Does Claude see all the feedback in our Enterpret workspace?
Claude sees what your own Enterpret account sees, nothing more. Each member connects with their own login, and the connector reaches one organization at a time. Within that scope it can query tickets, calls, reviews, surveys and Slack feedback that your workspace has brought in. A source that is not connected to Enterpret does not appear, and neither do its specific fields. If a colleague has wider access than you, their answers can differ from yours for the same question.
06Why do I see old tool names from Enterpret in Claude?
Because Claude keeps the tool list it received when you connected. Enterpret renamed its tools, and says the old names keep working for the rest of your current session. The new names show up once Claude refreshes the list, which usually means starting a new conversation. An old name that still shows simply belongs to the list loaded when the connection was first made. You never have to type any of them anyway: Enterpret notes that the assistant picks the right tool from what you ask. Nothing in your data changes during this switch.
07Should I use Claude or an automation tool with Enterpret?
They do different jobs, so it depends on the use. Claude with this connector suits questions you ask on the spot: what are users saying about this feature, give me three quotes, how many accounts raised it. An automation tool runs a fixed sequence on its own, such as a weekly summary sent to a channel. This connector never starts anything by itself, so scheduled digests belong elsewhere. The two can work side by side.