HashScraper API guide

How credits are used

By default, the bot's credits per item for every row. Each row of data a collection saves uses that bot's credits per item. By the basic calculation, a bot at 5 credits per item that collects 1,280 rows uses 6,400 credits. Some collection methods that save many rows in one batch can apply a multiplier, and accounts with a contract price use that price.
No rows, no credits. A collection that saves no data uses no credits. The reason is in the dataset's empty_reason.
Rows dropped as duplicates cost nothing. Rows judged to be duplicates and not saved use no credits. What counts as a duplicate depends on how the bot collects — saving the same content again in a new collection can use credits again.
Your balance is checked at the start. Without enough credits the collection does not start and you get 402 CREDIT_EXHAUSTED; with an expired service ticket, 402 TICKET_EXPIRED. A plan skips that run and records the reason in Plan runs — insufficient_credit for credits, ticket_expired and similar for service tickets.
The dataset has the real amount. The credits used by a dataset are in its credits_used (an integer). If a collection was appended to an earlier dataset with the stacking setting (union), this value is the total for everything stacked in that dataset, not just the latest collection.

curl 'https://www.hashscraper.com/v5/api/datasets/239758028' \
  -H 'Authorization: Bearer YOUR_API_KEY'

{
  "result": "success",
  "version": "v5",
  "request_id": "req_5a8c1f1c-9b2d-4e7c-9abf-3f7e0a4d1b21",
  "elapsed_sec": 0.0123,
  "data": {
    "id": 239758028,
    "bot": {
      "id": "8f3a7c1e9b5d24f6",
      "name": "Sample shopping bot"
    },
    "status": "done",
    "data_count": 1280,
    "union": false,
    "trigger": "api",
    "credits_used": 6400,
    "created_at": "2026-10-04T09:00:02+09:00",
    "updated_at": "2026-10-04T09:00:02+09:00",
    "plan": null,
    "inputs": {
      "param1": {
        "name": "Keyword",
        "value": "sneakers"
      }
    },
    "started_at": "2026-10-04T09:00:02+09:00",
    "elapsed_sec": 94
  }
}

{
  "result": "error",
  "version": "v5",
  "request_id": "req_5a8c1f1c-9b2d-4e7c-9abf-3f7e0a4d1b21",
  "code": "CREDIT_EXHAUSTED",
  "message": "Credits exhausted.",
  "elapsed_sec": 0.0011
}

Credits per item differ by bot

Credits per item differ by bot. Light data such as lists and comments costs less; detailed data or data with AI analysis costs more.

Prices can change, so this guide does not list numbers. These two places always show the current value.

Bot catalog — see the credits per item of every public bot without signing up.
GET /v5/api/bots, GET /v5/api/bots/{bot_id} — credit_per_item (an integer) in the response.

curl 'https://www.hashscraper.com/v5/api/bots/8f3a7c1e9b5d24f6' \
  -H 'Authorization: Bearer YOUR_API_KEY'

{
  "result": "success",
  "version": "v5",
  "request_id": "req_5a8c1f1c-9b2d-4e7c-9abf-3f7e0a4d1b21",
  "elapsed_sec": 0.0123,
  "data": {
    "id": "8f3a7c1e9b5d24f6",
    "name": "Sample shopping bot",
    "description": "Collects product lists by keyword.",
    "schedule_group": null,
    "union": false,
    "created_at": "2026-08-12T10:21:05+09:00",
    "updated_at": "2026-10-01T18:02:40+09:00",
    "worker_source": {
      "name": "Shopping list",
      "version": 12,
      "updated_at": "2026-09-20T11:00:00+09:00",
      "price": 5
    },
    "credit_per_item": 5,
    "plans_count": 2,
    "active_plans": [
      {
        "id": "512",
        "name": "주기 2",
        "label": "매주 월요일 09:00 (KST)"
      }
    ],
    "period_summary": "주기 2 · 매주 월요일 09:00 (KST)",
    "next_run_at": "2026-10-05T09:00:00+09:00",
    "latest_dataset": {
      "id": 239758028,
      "status": "done",
      "data_count": 1280,
      "trigger": "period",
      "created_at": "2026-09-28T09:00:03+09:00"
    }
  }
}

Rate limits

Requests are accepted up to 150 per minute per IP address. The count is by the IP address the requests come from, not by API key — several programs on the same server share one count.

Above that you get HTTP 429. The body is the same JSON as other errors, with code RATE_LIMITED. The Retry-After response header holds the number of seconds left before you may send again — wait that long, then send again. When you resend a request that starts a collection, use the same idempotency key.

This number is a guideline, not a guarantee. A few more requests may get through at times, but do not rely on it. Checking a collection every 10 to 30 seconds is enough, and to run many bots at once use bulk collection instead of one call per bot.

{
  "result": "error",
  "version": "v5",
  "request_id": "req_5a8c1f1c-9b2d-4e7c-9abf-3f7e0a4d1b21",
  "code": "RATE_LIMITED",
  "message": "Rate limit exceeded.",
  "elapsed_sec": 0.0011
}

import time
import uuid
import requests


def request_with_backoff(method, url, headers=None, **kwargs):
    headers = dict(headers or {})
    if method != "GET":
        # one key per request — every retry of the same request reuses it
        headers.setdefault("Idempotency-Key", str(uuid.uuid4()))
    for attempt in range(8):
        res = requests.request(method, url, headers=headers, timeout=30, **kwargs)
        if res.status_code != 429 or attempt == 7:
            return res
        # a 429 carries Retry-After: the seconds left until the one-minute window resets
        retry_after = res.headers.get("Retry-After", "")
        time.sleep(int(retry_after) if retry_after.isdigit() else min(2 ** attempt, 30))

// Node.js 18+ (built-in fetch)
const { randomUUID } = require("node:crypto");
const sleep = (ms) => new Promise((resolve) => setTimeout(resolve, ms));

async function requestWithBackoff(method, url, { headers = {}, body } = {}) {
  const h = { ...headers };
  // one key per request — every retry of the same request reuses it
  if (method !== "GET") h["Idempotency-Key"] ??= randomUUID();
  if (body) h["Content-Type"] = "application/json";
  for (let attempt = 0; ; attempt++) {
    const res = await fetch(url, { method, headers: h, body: body ? JSON.stringify(body) : undefined });
    if (res.status !== 429 || attempt === 7) return res;
    // a 429 carries Retry-After: the seconds left until the one-minute window resets
    const wait = Number(res.headers.get("Retry-After")) || Math.min(2 ** attempt, 30);
    await sleep(wait * 1000);
  }
}

How much one request can handle

What	Limit	Details
One list page (`limit`)	Up to 1,000 items	Paging through lists
Bulk collection (`POST /collections/bulk`)	Up to 100 items	Collections
Bulk job (`POST /collections/bulk_jobs`)	Up to 5,000 items	Collections
Excel input file	Up to 20MB	Settings
Plan run lookup range	Up to 31 days	Plan runs

Credits & rate limits V5 (beta)

How credits are used

Credits per item differ by bot

Rate limits

How much one request can handle

Related guides