> ## Documentation Index > Fetch the complete documentation index at: https://docs.rippit.com/llms.txt > Use this file to discover all available pages before exploring further. # Kustomer > Technical reference for the Rippit Kustomer data integration: API key authorization, API endpoints, what a conversation becomes in Rippit, attachments, and sync behavior. A read-only API key integration that syncs Kustomer conversations, messages, notes, customers, and agents into Rippit on a schedule. For what applies to every integration, see [how data integrations work](https://docs.rippit.com/plg/data-integrations.md). For the security model, see [Data integration security](https://docs.rippit.com/security/data-integrations.md). Kustomer is in beta. It appears as a source only in workspaces where beta integrations are turned on. ## Connection summary | | | | - | - | | Authorization | A Kustomer API key, sent as `Authorization: Bearer ` | | Scope parameter | None. Access is set by the roles on the key | | Key requirement | The org.user and org.admin read roles, plus org.admin.user, org.admin.content, and org.tracking | | Writes to Kustomer | None. Every call is a `GET` | | Credentials shared with Rippit | The API key. Rippit stores it and uses it for every sync | | Credential check | The key is tested against Kustomer before the connection is created. Rejected credentials create no connection, and Rippit does not keep them | | Connection identity | Your Kustomer org, read from the key. One connection per org | | Required Rippit role | Workspace admin | | API base | `https://.api.kustomerapp.com` | | API version | `v1`, in the path | | Sync method | Scheduled polling | | Attachments | Downloaded and stored by Rippit, including images inline in message and note bodies | | Maximum historical backfill | 90 days, which is also the default | | Field selection limit | 150 columns | ## Before you connect Rippit asks for two values. The subdomain is your org name in your Kustomer URL: `acme` in `acme.kustomerapp.com`. You can paste the whole URL, including an `acme.api.kustomerapp.com` URL, and Rippit extracts the subdomain. Create the API key in Kustomer under **Settings**, then **Security**, then **API Keys**. Give it the org.user and org.admin read roles, plus org.admin.user, org.admin.content, and org.tracking. Rippit reads your org from the key itself, so the key must belong to the org the subdomain names. ## Data handling | Question | Answer | | - | - | | Is conversation content ingested, or only metadata? | Full content: messages, notes, satisfaction ratings and comments, handle time, tags, custom attributes you select, customers, and agents | | Are notes ingested? | Yes. Notes are stored and marked as not public | | Are attachments stored? | Yes. File attachments and inline images are downloaded and stored rather than linked | | Can Rippit modify Kustomer? | No. No write code path exists | | Is data used to train models? | No. Rippit does not train any AI models. Your data is used for inference only | ### What a conversation becomes in Rippit Each Kustomer conversation becomes one conversation in Rippit. | Rippit record | Content | | - | - | | Subject | The conversation name | | Agents | Every assigned user, the user who sent the first response, and the agents who wrote messages or notes | | Start and end | The conversation's created time and its last updated time | | Status | The conversation status from Kustomer | | Tags | Tag names, looked up from your org's tags | | Default selected fields | Subject, status, tags, all assignees, commenters, first commenter, and last commenter | | Messages | Each message, attributed to the agent who sent it when outbound, or to the customer when inbound. Automated outbound messages are attributed to a bot. The text comes from the message's `preview` field | | Notes | Each note's body, attributed to its author and marked as not public | | Shortcuts | When a message uses a shortcut, Rippit records the shortcut's name, its saved text, and the message as sent | | Handle time | The conversation's handle time from Kustomer, added as a private system comment | | Satisfaction ratings | The rating, on a 1 to 5 scale, with any written answer added as a comment | | Attachments | Files attached to messages, and images inline in message and note bodies | | People | Agents, with name, display name, email, phone, and roles, and customers, with name, display name, and first phone number | The `categoryTree` and `contactReasonTree` custom attributes are synced as their display names rather than IDs. For tree attributes with several levels, the names are joined with commas. The field picker lists the attributes on recent conversations plus every conversation custom attribute defined in your org. Customer and agent fields, including custom attributes, are offered the same way. Each one is synced only if you select it. Conversations, messages, and notes deleted in Kustomer are marked as deleted in Rippit. ## API endpoints ### Authentication | Purpose | Endpoint | Method | | - | - | - | | Key check, which also reads the org the key belongs to | `/v1/auth/tokens/current` | GET | | Credential check, conversation access | `/v1/conversations` | GET | ### Conversations, messages, and people | Object | Endpoint | Method | | - | - | - | | Conversations changed since the last sync | `/v1/conversations?pageSize=1000&sort=desc` | GET | | Conversations, by ID | `/v1/conversations/` | GET | | A conversation's handle time | `/v1/conversations//times` | GET | | Messages changed since the last sync | `/v1/messages?pageSize=1000&sort=desc` | GET | | A conversation's messages, to re-sync it | `/v1/conversations//messages` | GET | | Notes changed since the last sync | `/v1/notes?pageSize=1000&sort=desc` | GET | | A conversation's notes, to re-sync it | `/v1/conversations//notes` | GET | | A message's attachments | `/v1/messages//attachments` | GET | | Attachments, by ID, to refresh a download URL | `/v1/attachments/` | GET | | All agents | `/v1/users?userType=user&pending=false` | GET | | Agents, by ID | `/v1/users/` | GET | | Customers, by ID | `/v1/customers/` | GET | | Tags | `/v1/tags` | GET | | Shortcuts | `/v1/shortcuts` | GET | | Conversation custom attribute definitions | `/v1/metadata/conversation` | GET | | Sample records and attribute definitions, for field discovery | `/v1/conversations`, `/v1/customers`, `/v1/users`, `/v1/metadata/customer`, and `/v1/metadata/user` | GET | | Attachment files | The `links.related` URL on each attachment, as returned by Kustomer | GET | The list endpoints page with the `links.next` value in each response, and Rippit follows it until it is absent. Customers and agents are not read as a full list for each sync. Rippit fetches the customer on each message, and the agents assigned to or first responding on each conversation, by ID. Requests by ID send a comma-separated list of up to 100 IDs, or 99 for customers. Kustomer attachment URLs expire within minutes, so before downloading each file Rippit refetches the attachment by ID to get a fresh URL. ## Sync behavior | | | | - | - | | Conversation discovery | Newest first by last updated time, read back until reaching records already synced | | Messages and notes | Their own lists, read the same way, so a new message or note is picked up without refetching its conversation | | Trailing buffer | Each sync reads up to 5 minutes before the current time | | Handle time | Fetched for each conversation listed in a sync | | Customers | Fetched by ID as messages reference them. Customers missing from conversations synced in the last 7 days are fetched by ID once a day | | Agents, tags, shortcuts, and custom attribute definitions | Full list on every sync | | Attachments | Downloaded after their message syncs. Downloads that failed in the last 5 days are retried once a day | | Update handling | An updated conversation, message, or note is listed again and replaced with the latest copy | | Resumption | High water mark on last updated time. Changed records only, no full rescan | ## Rate limiting Kustomer limits API calls for each org. Rippit reads `x-ratelimit-limit` and `x-ratelimit-remaining` on each response and slows down as the remaining budget runs low. Requests by ID run up to 10 at a time. This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.