Pagination

All list endpoints in the Organization and Enterprise APIs use cursor-based pagination. This provides consistent, efficient pagination regardless of the size of the result set.

How it works

Every list endpoint accepts two query parameters:

Parameter	Type	Description
`first`	integer	Maximum number of items to return per page (default varies by endpoint)
`after`	string	Opaque cursor from a previous response. Omit for the first page

Response format

List responses include pagination metadata:

{
  "items": [...],
  "has_next_page": true,
  "end_cursor": "eyJsYXN0X2lkIjoiYWJjMTIzIn0=",
  "total": 142
}

Field	Description
`items`	Array of results for the current page
`has_next_page`	`true` if there are more results
`end_cursor`	Pass this as the `after` parameter to get the next page. `null` when `has_next_page` is `false`
`total`	Total number of matching items (may be omitted by some endpoints for performance)

Example: Paginating through sessions

First page

curl "https://api.devin.ai/v3/organizations/$DEVIN_ORG_ID/sessions?first=10" \
  -H "Authorization: Bearer $DEVIN_API_KEY"

Use the end_cursor value from the previous response:

curl "https://api.devin.ai/v3/organizations/$DEVIN_ORG_ID/sessions?first=10&after=eyJsYXN0X2lkIjoiYWJjMTIzIn0=" \
  -H "Authorization: Bearer $DEVIN_API_KEY"

Collecting all results

import os
import requests

org_id = os.environ["DEVIN_ORG_ID"]
url = f"https://api.devin.ai/v3/organizations/{org_id}/sessions"
headers = {"Authorization": f"Bearer {os.environ['DEVIN_API_KEY']}"}
all_sessions = []
cursor = None

while True:
    params = {"first": 50}
    if cursor:
        params["after"] = cursor

    response = requests.get(url, headers=headers, params=params).json()
    all_sessions.extend(response["items"])

    if not response.get("has_next_page"):
        break
    cursor = response["end_cursor"]

print(f"Fetched {len(all_sessions)} sessions")

Migrating from offset-based pagination

If you’re migrating from API v1 or v2, replace offset/limit with after/first:

# Before (v1/v2)
curl ".../v1/sessions?offset=50&limit=25"

# After
curl ".../v3/organizations/{org_id}/sessions?first=25&after=CURSOR_FROM_PREVIOUS_PAGE"

Cursor-based pagination is more reliable than offset-based pagination because it isn’t affected by items being added or removed between pages.

Getting Started

Concepts

Agent Lifecycle

Context Management

Environment Management

Automation

Integrations

Analytics

Observability

Devin for Security

Infrastructure

Identity & Access

Configuration

Legacy API v1

Legacy API v2

How it works

Response format

Example: Paginating through sessions

First page

Next page

Collecting all results

​How it works

​Response format

​Example: Paginating through sessions

​First page

​Next page

​Collecting all results

​Migrating from offset-based pagination

How it works

Response format

Example: Paginating through sessions

First page

Next page

Collecting all results

Migrating from offset-based pagination