Endchan Scraper — Posts, Replies & Quote Links

Collect public Endchan posts and replies from board or thread URLs, with timestamps, source links and quote references. Export JSON, CSV or Excel for conversation research and repeated community snapshots. No Endchan login, source API key or proxy purchase is required.

Each run collects a bounded snapshot of the public boards or threads you supply. You control the maximum number of saved posts and can set an Apify spending limit.

Use cases: Endchan conversation research

Save one structured result for each original post or reply. Study discussions, follow supplied threads or compare board samples between runs. Stable post IDs connect records; anonymous names are not persistent identities.

Board pages discover threads; text comes from full public thread pages. It collects text and metadata without downloading media. Source policy and collection limits apply.

How to scrape Endchan posts and replies

1. Paste one or more public https://endchan.net/ board or thread URLs.
2. Choose Maximum posts, then start the run. The default is 20.
3. Open Posts and replies to inspect or export results. Check Collection summary to understand coverage and why the run stopped.

Replies are included automatically. Expand Optional settings to turn them off or apply filters.

Inputs and useful output

Input	How to use it
Board or thread URLs	Supply 1–10 public Endchan board, board-page or full-thread URLs.
Maximum posts	Choose 1–1,000. Originals and replies each count as one result.
Include replies	Enabled by default under Optional settings.
Post text keywords	Optional, case-insensitive OR matches over fetched posts. An original can also match its subject.
Published on or after / before	Optional inclusive UTC dates, written as YYYY-MM-DD. Posts without dates are excluded when date bounds apply.

The maximum is a ceiling. Filters apply to fetched posts, without global search or automatic ancestor retrieval. Direct thread URLs take priority over board discovery.

{"startUrls": ["https://endchan.net/tech/"], "maxItems": 20}

These abbreviated, sanitized records were observed in a public technology thread on 22 October 2026:

[
  {
    "recordId": "endchan.net:tech:15907",
    "recordType": "original",
    "text": "can someone help me by using handbrake to split an mkv into multiple mp4s and upload? is this the right place to ask?",
    "publishedAt": "2026-10-04T14:55:00.000Z",
    "quotedPostIds": [],
    "postUrl": "https://endchan.net/tech/res/15907.html#15907"
  },
  {
    "recordId": "endchan.net:tech:15939",
    "recordType": "reply",
    "text": ">>15907\nUse losslesscut",
    "publishedAt": "2026-10-20T02:24:00.000Z",
    "quotedPostIds": [
      "tech:15907"
    ],
    "postUrl": "https://endchan.net/tech/res/15907.html#15939"
  }
]

The default table highlights board, post type, thread title, text, publication time, quotes and source link. Full JSON adds author, collection/edit times, outbound links and structured quote targets. Unavailable values remain null; thread titles can be absent.

Use recordId to join snapshots. Quote IDs include the board, because numeric IDs can repeat. Quotes do not establish parents. Choose Snapshot comparison to compare contentHash for the same record ID. Hashes describe normalized, redacted output; they do not prove edit history. Older saved rows may lack a hash.

How much does Endchan scraping cost?

$1 per 1,000 saved posts/replies ($0.001 each), plus $0.02 per started run. Run compute and source fetching are included; there is no subscription.

Saved posts/replies in one run	Total charge
0	$0.02
20	$0.04
100	$0.12
1,000	$1.02

These totals assume enough accessible content fits within the collection limits. The Pricing tab is authoritative.

The start fee also applies to empty or unavailable-source runs. Previews, filtered posts, within-run duplicates, errors and summary records have no result fee. Repeated runs are new snapshots and can save the same posts again. Ordinary Apify storage and exports after the run can incur platform charges. A minimum spending limit of $0.021 covers the start plus at most one post; it does not guarantee a result. A $0.04 limit permits up to 20 saved posts. Multiple small runs each incur a start fee.

Coverage and common questions

What does the collection status mean?

The summary starts with a short explanation: finished means the bounded scope was scanned; limited means a cap or pending work remains; partial means a source omission or page failure; failed means collection failed; interrupted means shutdown retained saved work. Zero matching results can be a valid finished scan.

Can it collect every reply or all history?

It collects accessible replies in fetched HTML within 10 content attempts by default (including retries; up to 20 with an explicit maxRequests), a two-minute source-request deadline, 2 MiB per response and 8 MiB total download reservations. Startup, parsing and storage can add time; use a 150-second host timeout. Missing, deleted, private and omitted historical content is not recovered.

What if access is denied or fewer posts appear?

Check the summary for filters, limits and failed URLs. The Actor respects source policy and stops for access refusals or rate limits. It does not bypass them. For support, provide the run ID and public URL through the Issues tab when available; never send credentials.

Public access does not grant commercial reuse rights. Confirm permissions for your intended use. Common contact patterns are redacted, but this is not a complete personal-data detector. This independent tool is not endorsed by Endchan.

Use Endchan Scraper with Python

Uses apify-client 3.x and an Apify token, without Endchan credentials. The maximum-charge setting covers the run start and saved-result fees. This example permits up to 20 results for $0.04.

import os
from datetime import timedelta
from decimal import Decimal
from apify_client import ApifyClient

client = ApifyClient(os.environ["APIFY_TOKEN"])
run = client.actor("thescrapelab/endchan-public-discussion-scraper").call(
    run_input={"startUrls": ["https://endchan.net/tech/"], "maxItems": 20},
    memory_mbytes=256,
    run_timeout=timedelta(seconds=150),
    max_total_charge_usd=Decimal("0.04"),
    restart_on_error=False,
)
if run is None:
    raise RuntimeError("The Actor did not return a run")
for post in client.dataset(run.default_dataset_id).iterate_items():
    print(post["recordId"], post["text"], post["postUrl"])