Integration · CRM & sales

Connect Claude & ChatGPT to Unify

Ask your AI to pull a list of companies, check whether a person already exists, or update a record in Unify. It reads your objects and their attributes first, so it knows which fields exist. It can run record and event queries, find records by unique values like a domain or email, and create, update or upsert records.

Connect Unify free

Free plan, no credit card. Takes about a minute.

Try asking

“Which objects and fields do we have in Unify?”
The AI lists your objects, such as company, person and opportunity, and the attributes of the one you ask about.
“Is acme.com already a company in Unify?”
It finds the company record by its unique domain value and shows its attribute values, or tells you there is no match.
“Add Jane Miller, [email protected], as a person linked to Acme.”
It checks the attributes, looks for an existing record by email and creates the person with the values you gave.

What your AI can do

Objects and attributes

List objects, their attributes and the allowed options of select fields, so the AI writes values that fit your setup.

Records

Find, read, create, update and upsert records, and delete a record if you switch that tool on.

Record and event queries

Run query jobs over records or tracked events with filters and sorting, then read the paginated results.

“List companies in Denmark, sorted by name.”
It starts a record query with the attributes and filter you describe, waits for the job and reads the results.
“Set the status on opportunity 4821 to Negotiation.”
It checks the allowed options for the status attribute and updates only that field on the record.
“What events did we track for acme.com last week?”
It starts an event query with a filter and reads the events from the finished job.

17 tools for Unify

These tools are switched on when you connect. You can switch any of them off, or require your approval before it runs.

create_recordChanges dataCreate a record of an object (`object_name`) with attribute values in `data`. All required attributes must be included. Find the record first to avoid duplicates.
delete_recordDeletesPermanently delete a record by `object_name` and `record_id`. This cannot be undone.
find_recordChanges dataFind a single record of an object by unique attribute values in `match` (e.g. a domain or email). Use this to check if a record exists before creating one.
get_attributeGet one attribute of an object by `object_name` and `attribute_name`, including its type and settings.
get_event_queryCheck the status of an event query job by `job_id`.
get_event_query_resultsGet the events of a finished event query job by `job_id`. Results are paginated.
get_objectGet one object's definition by its API name (`project_name`), e.g. company or person.
get_recordGet one record by `object_name` and `record_id`, with its attribute values.
get_record_queryCheck the status of a record query job by `object_name` and `job_id`.
get_record_query_resultsGet the matching records of a finished record query job by `object_name` and `job_id`. Results are paginated.
list_attribute_optionsList the allowed options of a select-type attribute, by `object_name` and `attribute_name`.
list_attributesList the attributes (fields) of an object by `object_name`. Use this to learn which fields you can read, filter on or write.
list_objectsList all objects in the Unify workspace (e.g. company, person, opportunity and custom objects) with their API names. Start here to learn which objects exist.
start_event_queryChanges dataStart a query job over tracked events, with an optional `filter`. Returns a job ID; fetch rows with get_event_query_results.
start_record_queryChanges dataStart a query job for records of an object (`object_name`). The `query` sets which attributes to `select` and optional `where` filters and sorting. Returns a job ID; fetch rows with get_record_query_results.
update_recordChanges dataUpdate a record's attribute values in `data`, by `object_name` and `record_id`. Only the attributes you pass change.
upsert_recordChanges dataCreate or update a record of an object: `match` identifies an existing record by unique attribute values, and `create` / `update` hold the values to write in each case.

Set up in three steps

1
Pick the app
Create a free PipMCP account and choose Unify from the app list.
2
Paste your key
Paste a Unify API key. In Unify go to Settings › Developers (API keys) and create an API key. You need to be an admin to do this.
3
Add the link to your AI
You get a personal MCP link. Add it to Claude, ChatGPT or Cursor:
1. Click your name, then Settings › Connectors › Add custom connector.
2. Paste your link as the Remote MCP server URL.
3. Switch it on from the + menu in a chat.

Questions

What can the AI do in Unify?

The AI can list objects, attributes and attribute options, run query jobs over records and tracked events, and find or read single records. It can create, update and upsert records, and delete a record if you switch that tool on.

Does the AI see my Unify API key?

No. Your Unify API key are encrypted at rest and never shown to the AI. After you save them, they are not shown again, not even to you. The AI only sees the results of the tools it calls.

Can I control what the AI is allowed to do?

Yes. You choose which tools are switched on, so you can start with lookups and queries only. Deleting a record is permanent and can require your approval before it runs, and so can creating or updating records. Every tool call is logged.

Does it work with ChatGPT?

Yes. In ChatGPT go to Settings › Apps & Connectors › Advanced and turn on Developer mode, then add your PipMCP link. Developer mode needs a paid ChatGPT plan: Plus, Pro, Business or Enterprise. The same link also works in Claude (Settings › Connectors › Add custom connector), Cursor and other MCP clients.

What does it cost?

PipMCP has a free plan with no credit card required. Paid plans bill per completed task. You also need your own Unify account.

Why does the AI start with objects and attributes?

Every Unify workspace can have its own objects and fields, including custom ones. By listing objects, attributes and select options first, the AI uses the right API names and allowed values when it queries or writes records, instead of guessing.