README.md

metadata

license: cc-by-sa-3.0
language:
  - en
pretty_name: Secure, Contain, Protect
tags:
  - scp
  - scp-foundation
  - creative-writing
  - fiction
  - horror
  - wiki
size_categories:
  - 10K<n<100K
task_categories:
  - text-generation
  - text-classification
  - fill-mask
  - text-retrieval
  - sentence-similarity
  - feature-extraction
  - tabular-regression
  - text-ranking
configs:
  - config_name: pages
    data_files: data/parquet/pages.parquet
    default: true
  - config_name: page_tags
    data_files: data/parquet/page_tags.parquet
  - config_name: page_alternate_titles
    data_files: data/parquet/page_alternate_titles.parquet

Secure, Contain, Protect

A structured snapshot of the English SCP Foundation Wiki — every content page (SCPs, tales, hubs, GOI formats, essays and art) created between 2008 and 2026, harvested from the Crom GraphQL API. 19,438 pages by 3,479 credited authors, each with its raw Wikidot source text, rating, vote and comment counts, tags, authorship, alternate titles, and per-credit attributions (the credited author(s), distinct from whoever posted the page).

User/author pages, hidden pages, and non-content categories are filtered out; only the rendered HTML is omitted (the raw source is included so you can render or parse it yourself). Snapshot date: 2026-06-03.

Dataset structure

Three tables, published as one Parquet file each — flat columns throughout, except pages.attributions, which is a JSON array of credits.

pages — one row per content page (19,438 rows):

column	type	notes
url	string	canonical Wikidot URL (primary key)
wikidot_id	string	Wikidot's internal page id
title	string
rating	double	net votes (nullable)
vote_count	int
category	string	always _default (content pages)
created_at	timestamp	UTC
revision_count	int
comment_count	int
thumbnail_url	string	nullable
parent_url	string	parent page, if any (nullable)
created_by_crom_id	string	null if the author account was deleted
created_by_display_name	string
created_by_unix_name	string
created_by_wikidot_id	string
source	string	raw Wikidot markup
summary	string	author-provided summary (often null)
attributions	json	array of credits {type, user_display_name, date, order}; type ∈ AUTHOR / REWRITE / TRANSLATOR / SUBMITTER
fetched_at	timestamp	UTC; when the row was crawled

page_tags — (page_url, position, tag), 175,007 rows.

page_alternate_titles — (page_url, idx, title, source), 10,465 rows.

Join the children to pages on page_url = url.

Use the published dataset

Pull it straight from the Hub — no local build required.

# 🤗 datasets
from datasets import load_dataset

pages = load_dataset("ozefe/secure-contain-protect", "pages", split="train")
tags  = load_dataset("ozefe/secure-contain-protect", "page_tags", split="train")

# pandas / polars, read directly from the Hub
import pandas as pd

pages = pd.read_parquet("hf://datasets/ozefe/secure-contain-protect/data/parquet/pages.parquet")

-- DuckDB, query the Hub without downloading
SELECT title, rating
FROM 'hf://datasets/ozefe/secure-contain-protect/data/parquet/pages.parquet'
ORDER BY rating DESC
LIMIT 10;

# Or download every file locally
huggingface-cli download ozefe/secure-contain-protect --repo-type dataset --local-dir scp-data

A quick join — the most common tags on the highest-rated SCPs:

import duckdb

con = duckdb.connect()
base = "hf://datasets/ozefe/secure-contain-protect/data/parquet"
print(con.execute(f"""
    SELECT t.tag, count(*) n
    FROM '{base}/pages.parquet' p
    JOIN '{base}/page_tags.parquet' t ON t.page_url = p.url
    WHERE p.rating > 1000
    GROUP BY t.tag ORDER BY n DESC LIMIT 10
""").fetchall())

The attributions column is a JSON array — unnest it to get the real credited authors (distinct from created_by_*, which is whoever posted the page):

SELECT title, a.user_display_name, a.type
FROM 'pages.parquet', unnest(from_json(attributions,
        '[{"user_display_name": "VARCHAR", "type": "VARCHAR", "date": "VARCHAR", "order": "INTEGER"}]')) AS t(a)
WHERE url = 'http://scp-wiki.wikidot.com/scp-682';
-- Dr Gears (AUTHOR), Epic Phail Spy (AUTHOR)

Build it yourself

The repository ships the full pipeline: crawl -> export -> report. Requires Python 3.14+ (the Crom GraphQL API client thaumiel needs it).

python -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
cd src

1. Download the data into DuckDB (../data/scp_dataset.duckdb). The crawl is per-year, resumable (the database is the checkpoint), and Ctrl+C-safe:

python scp_dataset.py 2008  # a single year
for y in $(seq 2008 2026); do python scp_dataset.py "$y"; done  # the whole archive

2. Export to Parquet — one file per table in ../data/parquet/:

python export_parquet.py

3. Statistics & plots — writes ../images/*.png and the STATISTICS.md report below:

python generate_report.py

Crom meters requests against a 300,000-point budget that resets every 5 minutes; one full year costs only a few thousand points, so the whole archive fits comfortably. The crawler logs the remaining quota as it goes.

How it was built

Pages are selected server-side from Crom GraphQL API with: URL on scp-wiki.wikidot.com, Wikidot category _default, is_hidden = false, and is_user_page = false. Per page the crawler requests the free fields plus the costly source, summary, alternate_titles, and attributions (the latter stored as a JSON column); it skips only text_content (rendered HTML). Timestamps are stored as UTC. See src/scp_dataset.py.

License & attribution

The SCP Foundation Wiki is licensed under Creative Commons Attribution-ShareAlike 3.0, and this dataset — which contains that source text — is released under the same license. If you use it, you must attribute the SCP Foundation Wiki and the individual authors (see the created_by_* columns and the license boxes embedded in each page's source), and share derivatives under CC BY-SA 3.0. Data accessed via the Crom GraphQL API. This dataset is an unofficial, fan-made compilation and is not affiliated with the SCP Wiki administration.

The SCP Foundation Wiki in Numbers

SCP Wiki content pages · 2008–2026 · n = 19,438 · via Crom.

A statistical tour of every content page (SCPs, tales, hubs, GOI formats and more) on the English SCP Wiki, built from the Crom API. Each figure below is followed by the conclusion it supports.

Content created per year

Takeaway. The wiki grew from 298 pages in 2008 to a peak around 2025, 19,438 in all. SCPs are the backbone but tales now make up roughly a third of yearly output; 2026 is a partial year.

Object classes

Takeaway. Euclid ('anomalous but containable') is the plurality at 3,431, just ahead of Safe (3,084); Keter is a distant third. The catch-all 'esoteric-class' has overtaken every named class except the big three.

Most common themes

Takeaway. Entity descriptors dominate — 'sapient', 'humanoid' and 'alive' lead — followed by genre tags like horror and mind-affecting. Comedy ranks surprisingly high, a reminder the wiki is not all grimdark.

Rating distribution

Takeaway. Ratings are extremely right-skewed: a median of 69 against a maximum of 10,758 (SCP-173). Only 46 pages sit below zero — the community deletes weak work, so what remains is heavily curated.

Article length

Takeaway. Article length is roughly log-normal around a median of 10,398 characters (a few pages of text), with a long tail of giant hubs and anthologies reaching 199,994 characters.

Authorship is a power law

Takeaway. Counting every credited author (not just whoever posted the page), contribution is steeply unequal: the most prolific 2% hold 30% of all author credits, while 1,830 authors have a single credit. A small core sustains the wiki.

Does length buy quality? No

Takeaway. Rating and article length are essentially uncorrelated (r = -0.01). The median rating is flat across the whole length range — a longer article is not a better-rated one. Concept and execution, not word count, drive a page's score.

Ratings fall with each cohort

Takeaway. Average rating slides from 511 in 2008 to 36 in 2026. This is mostly an age effect — older pages have had years to accumulate up-votes — rather than proof that newer writing is worse; the median falls far more gently than the mean.

Danger sells

Takeaway. The more dangerous the classification, the higher the average score: Apollyon (224) and Keter top the table, while Safe and Neutralized trail. Readers reward existential threat over the mundane.

Prolific vs. acclaimed authors

Takeaway. Counting every credited author (co-authors included), output and acclaim remain different games: the most prolific cluster at modest average ratings, while the highest-rated are comparatively selective — bubble size (total score) shows a few writers manage both volume and quality.

Co-authorship over time

Takeaway. Co-authorship has climbed from near zero to about 10% of pages in recent years — modern SCP is increasingly a team effort (2026 is a partial year). Early collaboration is undercounted: the attribution metadata recording co-authors is a later convention.

Bigger teams, higher ratings

Takeaway. Median rating rises with team size — from 68 for solo pages to 103 for the largest teams. Collaboration correlates with a warmer reception, though the most ambitious projects also tend to attract co-authors.

How themes cluster

Takeaway. Themes form clear clusters. The strongest pairing is 'sapient' + 'humanoid': entity descriptors (humanoid / sapient / alive) co-occur tightly, forming a 'monster' cluster distinct from the genre tags.

How authors collaborate

Takeaway. Co-authorship forms tight clusters around a few hubs — Ralliston is the most connected. Node size is total pages, edge weight is shared pages; the modern wiki is a densely woven collaborative network, not a crowd of soloists.

Per-year summary

Year	Pages	SCPs	Tales	Authors	Avg rating	Top-rated page
2008	298	237	24	50	511	SCP-173
2009	376	248	66	114	301	SCP-049
2010	393	206	132	127	255	SCP-096
2011	488	281	172	127	241	SCP-1000
2012	940	571	320	248	227	SCP-____-J
2013	744	391	296	163	205	SCP-2000
2014	780	332	328	168	182	SCP-2317
2015	622	332	239	177	220	●●|●●●●●|●●|●
2016	648	306	254	192	157	SCP-2316
2017	1,023	547	339	248	187	SCP-3008
2018	1,686	904	580	405	149	REDACTED PER PROTOCOL 4000-ESHU
2019	1,404	740	478	366	118	SCP-4205
2020	1,424	721	485	366	114	SCP-5000
2021	1,583	859	489	411	103	SCP-6001
2022	1,738	1,001	522	491	83	SCP-7000
2023	1,325	672	496	383	62	SCP-7819
2024	1,289	629	485	407	62	SCP-8980
2025	1,785	934	630	563	50	SCP-9000
2026	892	556	207	375	36	National Fog Safety Initiative

Takeaway. Output and contributor counts climb together — more authors, more pages — while average rating declines with each younger cohort (an age effect, not a quality one).

Top 15 highest-rated pages

#	Title	Rating	Votes	Comments	Author	Year
1	SCP-173	10,758	11,854	2,039	Lt Masipag	2008
2	●●|●●●●●|●●|●	7,150	7,316	613	LurkD	2015
3	SCP-049	5,547	6,035	700	Gabriel Jade	2009
4	SCP-____-J	5,247	5,717	742	Communism will win	2012
5	SCP-096	4,792	5,006	566	Dr Dan	2010
6	SCP-055	4,655	4,753	502	xthevilecorruptor	2008
7	SCP-682	4,210	5,056	1,057	Dr Gears	2008
8	SCP-5000	4,195	4,485	446	Tanhony	2020
9	SCP-087	3,817	3,997	466	Zaeyde	2009
10	SCP-3008	3,725	3,819	353	Mortos	2017
11	SCP-106	3,679	3,873	643	Dr Gears	2010
12	SCP-999	3,494	3,988	305	ProfSnider	2009
13	SCP-093	3,402	3,566	238	far2	2008
14	REDACTED PER PROTOCOL 4000-ESHU	3,359	3,569	441	PeppersGhost	2018
15	SCP-914	3,280	3,366	218	far2	2008

Takeaway. The canon is front-loaded with early classics — SCP-173, SCP-049, SCP-682 — that have compounded votes for over a decade, alongside a few modern breakouts like SCP-5000.

Top 15 authors (all credited authors)

#	Author	Pages	Avg rating	Total rating	Co-authored	Best-rated work
1	Uncle Nicolini	341	109	37,235	35%	SCP-5555
2	HarryBlank	251	148	37,031	23%	SCP-7000
3	RJB_R	200	112	22,450	0%	SCP-1833
4	Ralliston	182	93	16,858	45%	SCP-6747
5	Tanhony	160	260	41,537	4%	SCP-5000
6	djkaktus	152	445	67,646	13%	SCP-1730
7	Doctor Cimmerian	145	165	23,876	19%	Cimmerian-Kaktus Proposal
8	Communism will win	140	222	31,053	1%	SCP-____-J
9	Dr Gears	138	336	46,412	8%	SCP-682
10	Rounderhouse	135	235	31,732	22%	SCP-6000
11	DrClef	132	323	42,659	11%	SCP-2317
12	daveyoufool	130	212	27,508	3%	SCP-TTKU-J (which is a thing that kills you)
13	DarkStuff	128	103	13,120	21%	SCP-6500
14	A Random Day	119	165	19,631	13%	SCP-3000
15	Zyn	115	146	16,759	35%	SCP-348

Takeaway. Crediting every author (co-authors, and people who never posted their own work) reshuffles the leaderboard versus a naive by-poster count — and the co-authored share shows how collaboratively each writer works.

Top co-author duos

#	Author A	Author B	Shared pages	Avg rating
1	HarryBlank	Placeholder McD	25	180
2	Jasiu06	Ralliston	22	75
3	Ralliston	Trotskyeet	16	127
4	Ben Counter	Pacific Obadiah	13	36
5	J Dune	PlaguePJP	13	312
6	Grigori Karpin	HarryBlank	13	234
7	DrAkimoto	MalyceGraves	12	65
8	LORDXVNV	Ralliston	11	171
9	DarkStuff	Uncle Nicolini	10	116
10	JakdragonX	Ralliston	10	117
11	AnAnomalousWriter	Ecronak	9	273
12	Liryn	Placeholder McD	9	525

Takeaway. The wiki's tightest writing partnerships — recurring duos that have co-authored many pages together, several rating well above the site median.

Most collaborative authors

#	Author	Distinct co-authors	Co-authored pages
1	syuzhet	317	25
2	Elenee FishTruck	232	15
3	Uncle Nicolini	133	119
4	Ralliston	103	81
5	Lt Flops	95	38
6	LORDXVNV	82	28
7	Dino--Draws	78	22
8	Rhineriver	73	4
9	stormbreath	72	16
10	PeppersGhost	70	12
11	kura_art	70	2
12	Pedagon	70	16
13	Prismal	64	21
14	Elenee Fishtruck	62	3
15	HarryBlank	62	57

Takeaway. The community's connectors — authors who have written with the widest circle of collaborators, knitting otherwise separate clusters together.

Top rewriters

#	Rewriter	Pages rewritten	Avg rating
1	Voct	30	303
2	Uncle Nicolini	13	94
3	Communism will win	10	329
4	thedeadlymoose	6	661
5	Queerious	5	58
6	Drewbear	5	309
7	DrClef	5	1085
8	Quikngruvn	5	275
9	SimpleCadence	5	261
10	JakdragonX	5	133
11	Aelanna	5	236
12	Tstaffor	4	263

Takeaway. The canon's caretakers: a small group does most of the rewriting that keeps the early, heavily-trafficked articles current.