Skip to main content
GET
List retrieval requests

Authorizations

Authorization
string
header
required

API key authentication. Use format: Authorization: Bearer tavrn_sk_live_...

Query Parameters

limit
integer
default:20

Maximum number of results

Required range: 1 <= x <= 100
cursor
string

Pagination cursor

status
enum<string>

Filter by coarse retrieval status

Available options:
received,
processing,
action_required,
completed,
cancelled
stage
string

Filter by lifecycle stage. Accepts a single stage or a comma-separated list (e.g. request_sent,acknowledged). May be combined with status.

case_id
string

Filter by case ID

created_after
string<date-time>

Exclusive creation timestamp lower bound (ISO 8601). May be combined with created_before.

created_before
string<date-time>

Exclusive creation timestamp upper bound. May be combined with created_after.

public_updated_since
string<date-time>

Inclusive lower bound for retrieval changes included in incremental polling. Start with the epoch for an initial full sync, then use the prior completed round's next_public_updated_since. Do not derive it from your clock or event timestamps.

include
enum<string>

Include timeline, pending client action, and document metadata in each result. Without it, returns summaries.

Available options:
details

Response

Successful response

results
object[]
required
next
string<uri> | null
required
Example:

"https://enterprise.tavrn.ai/v2/retrieval-requests?limit=20&cursor=next_cursor"

previous
string<uri> | null
required
Example:

null

scan_until
string<date-time>

Upper time bound for this polling round. Present only with public_updated_since; do not use it as the next checkpoint.

Example:

"2026-10-08T12:00:00.000Z"

next_public_updated_since
string<date-time>

Checkpoint returned only on the final incremental page, including an empty final page. Save after processing every page and use it as the next public_updated_since. Results can repeat across rounds.

Example:

"2026-10-08T11:58:00.000Z"