> ## 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. # Sierra > Technical reference for the Rippit Sierra data integration: API token authorization, API endpoints, what a conversation becomes in Rippit, and sync behavior. A read-only API token integration that syncs Sierra AI agent conversations into Rippit on a schedule. Sierra is in beta. 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). ## Connection summary | | | | - | - | | Authorization | Sierra API token, sent as `Authorization: Bearer ` | | Scope parameter | None | | Writes to Sierra | None. Every call is a `GET` | | Credentials shared with Rippit | The API token, your organization ID, and your agent ID. Rippit stores them and uses them for every sync | | Credential check | The token is tested against Sierra before the connection is saved. A rejected token creates no connection, and Rippit does not keep it | | Connection identity | Your Sierra organization ID. One connection per organization ID. Organization IDs are not case-sensitive, so `AcmeOrg` and `acmeorg` are the same connection | | Agents per connection | One. The connection reads conversations from the agent ID you enter | | Required Rippit role | Workspace admin | | API base | `https://api.sierra.ai/admin/1/orgs//agents/` | | API version | `1`, in the path | | Sync method | Scheduled polling | | Voice recordings | Not downloaded. Rippit stores the text of each message | | Historical backfill | 90 days by default, and up to 90 days, counted back from the start of today (UTC) by each conversation's start time | | Field selection limit | 150 fields at setup | ## Before you connect Rippit asks for three values: an API token, your organization ID, and your agent ID. You can find all three on your Sierra docs site. The organization ID and agent ID accept letters, numbers, underscores, and hyphens. The token must be able to read conversations and messages for that agent. Rippit uses it only to read. ## Data handling | Question | Answer | | - | - | | Is conversation content ingested, or only metadata? | Full content: every message, plus the conversation's duration, message count, bot ID, and tags. Rippit stores each conversation exactly as Sierra returns it, including attributes it does not offer as fields | | Are voice recordings stored? | No | | Can Rippit modify Sierra? | No. No write code path exists | ### What a conversation becomes in Rippit Each Sierra conversation becomes one conversation in Rippit. | Rippit record | Content | | - | - | | Subject | The Sierra conversation ID | | Agent | One agent per connection, named `Sierra AI`, who authors every AI message | | Start and end | The conversation's start time, and the time of its last message | | Fields | Subject, source, tags, duration, message count, bot ID, and the commenter fields | | Default selected fields | Subject, duration, message count, tags, commenters, first commenter, and last commenter | | Message comments | Every message, in order. AI messages are attributed to the Sierra AI agent. User messages are attributed to the customer, without a name or email. Messages from a human agent are kept, but are not linked to a named person | The fields above are the full list offered during setup. Rippit does not add other attributes from your Sierra conversations to the list. ## API endpoints | Object | Endpoint | Method | | - | - | - | | Credential check when you connect | `/conversations/export?start=<2 hours ago>&end=<1 hour ago>` | GET | | Conversations | `/conversations/export?start=&end=&limit=1000` | GET | | Next page of conversations | `/conversations/export?cursor=&limit=1000` | GET | | Field sample | `/conversations/export?start=<1 week ago>&end=<1 hour ago>`, the first conversation returned | GET | All paths are relative to the API base above. `/conversations/export` returns each conversation in full, with its messages, so no other endpoint is called for content. Timestamps are Unix seconds. Sierra pages with a `next_cursor` value, and Rippit requests the next page with it until none is returned. ## Sync behavior | | | | - | - | | Conversation discovery | Windowed on the conversation's start time, cursor pagination | | Window width | 1 hour, one window at a time | | Trailing buffer | Each window closes 1 hour behind the current time | | Deletions | A conversation deleted in Sierra stays in Rippit | This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.