V1 to V2 Endpoint Mapping

The V1-era public API documented 13 operations: ten under /api/v1 and three public operations without a version prefix. The table below maps all 13 to V2. V2 exposes 34 public operations in total; the other 21 are new capabilities without a direct V1 equivalent. A single V1 body often becomes a combination of a V2 path, query string, and smaller JSON body.

Complete endpoint comparison

V1 operation	V2 replacement	Required changes
GET /api/scraper	GET /api/v2/workers/{workerId} and GET /api/v2/workers/{workerId}/input-schema	Move the slug query value to {workerId} and encode owner/name as owner~name. Authentication is now required for Worker detail; read the public input-schema endpoint when only the input contract is needed.
GET /api/store	GET /api/v2/store	Change search to keyword; add zero-based offset when paging. limit now defaults to 20 and cannot exceed 100.
GET /api/proxy/region	GET /api/v2/proxy/region	Add the /v2 path segment. The optional language query parameter is available in V2.
POST /api/v1/run/result/list	GET /api/v2/worker-runs/{runId}/result	Move run_slug to {runId}. Convert page_index and page_size to offset and limit query parameters.
POST /api/v1/run/detail	GET /api/v2/worker-runs/{runId}	Move run_slug from the body to {runId}. Remove the JSON body.
POST /api/v1/scraper/run	POST /api/v2/workers/{workerId}/runs	Replace scraper_slug with a verified {workerId} path value. Keep version, input, callback_url, and is_async in the body. Rename page_index/page_size to offset/limit.
POST /api/v1/rerun	POST /api/v2/worker-runs/{runId}/rerun	Move run_slug to {runId}. Keep callback_url in the body; V2 also supports is_async, offset, and limit.
POST /api/v1/run/result/export	GET /api/v2/worker-runs/{runId}/result/export	Move run_slug to {runId}. Send format and comma-separated filter_keys as query parameters. Remove the JSON body.
POST /api/v1/task/run	POST /api/v2/worker-tasks/{workerTaskId}/runs	Move task_slug to {workerTaskId}. Keep callback_url in the body; V2 also supports is_async, offset, and limit.
POST /api/v1/run/list	GET /api/v2/worker-runs	Move filters to query parameters. Rename scraper_slug to worker_id; convert pagination and numeric status values. Omit status to list all statuses.
POST /api/v1/run/last/log	GET /api/v2/worker-runs/{runId}/log	Despite the V1 path name, the body selected a run by run_slug. Move it to {runId} and remove the body.
POST /api/v1/scraper/abort	POST /api/v2/worker-runs/{runId}/abort	Move run_slug to {runId} and remove the body. V2 returns the standard response envelope.
POST /api/v1/account/info	GET /api/v2/users/account	Change POST to GET and remove the empty JSON body. Account response fields changed; update deserialization.

Field and behavior changes

Area	V1	V2	Migration action
Authentication	api-key header	Bearer recommended; legacy header and query token also supported	Send Authorization: Bearer YOUR_API_KEY.
Resource name	scraper in request concepts	worker in paths and request concepts	Rename client models and variables, but keep documented scraper_* response fields.
Identifiers	IDs in JSON bodies	workerId, workerTaskId, and runId in paths	URL-encode path values and verify the V2 Worker identifier.
Pagination	page_index starts at 1; page_size	offset starts at 0; limit is at most 100	Use offset = (page_index - 1) * page_size.
Run status	Integers 1 through 5; 0 means all in a filter	Strings ready, running, succeeded, failed, aborting	Replace numeric comparisons; omit an all-status filter.
Result export fields	filter_keys JSON array	Comma-separated filter_keys query string	Join field names with commas and URL-encode the query.
Export formats	csv, json documented	csv, json, jsonl, xlsx, xls, xml, html, rss	Keep an allowed lowercase value; do not rely on V1 validation.
Response envelope	code, message, data	code, message, request_id, and usually data	Add request_id to logs and allow success responses without data where documented.
Validation	Mainly 400, 401, 429 documented	Also uses 404, 422, and standardized error envelopes	Branch on HTTP status and application code.
Account data	balance, traffic, traffic_expiration_at	Current contract example exposes balance, balance_expiration_at	Do not require removed traffic fields; follow the account endpoint schema.
Run logs	Log list items were structured objects in V1	Current V2 example exposes the log list as strings	Treat data.list according to the V2 endpoint contract, not a V1 log-item model.

Request conversions

List results

// V1 body
{ "run_slug": "RUN_ID", "page_index": 3, "page_size": 20 }

GET /api/v2/worker-runs/RUN_ID/result?offset=40&limit=20

Filter runs

// V1 body: status 3 meant succeeded
{ "page_index": 1, "page_size": 20, "status": 3, "scraper_slug": "WORKER_ID" }

GET /api/v2/worker-runs?offset=0&limit=20&status=succeeded&worker_id=WORKER_ID

Export selected fields

// V1 body
{ "run_slug": "RUN_ID", "format": "csv", "filter_keys": ["title", "address"] }

GET /api/v2/worker-runs/RUN_ID/result/export?format=csv&filter_keys=title%2Caddress

V2 capabilities without a direct V1 equivalent

V2 also exposes Worker discovery and input schemas, Worker task CRUD, account-level and Worker-level latest-run shortcuts, and dedicated endpoints for run logs, results, exports, cancellation, and reruns. Use the public endpoint reference to avoid rebuilding these capabilities from V1-era workarounds.