SMMScout Open Data — Free SMM Panel Dataset & API

The SMMScout dataset is a free, open collection of SMM panel data: 129 SMM panels with Scout Scores, measured API latency, service counts, payment methods, verification status and dated flags — served as JSON and CSV and refreshed automatically every week. It answers the question most panel rankings avoid: what are the measured numbers behind each panel, on what date were they taken, and can the ranking be recomputed from the published formula.

Use it to research panels, build price monitors, power comparison tools or train models. No API key. CORS open. CC BY 4.0.

Fetch with curl + jq — the 10 panels with the most services:

curl -s https://smmscout.com/api/panels.json \
  | jq -r '.panels | sort_by(.services // 0) | reverse | .[:10][] | "\(.name)\t\(.services)"'

Fetch in JavaScript (browser or Node):

const res = await fetch("https://smmscout.com/api/panels.json");
const data = await res.json();
const fastest = [...data.panels]
  .filter((p) => p.response_ms)
  .sort((a, b) => a.response_ms - b.response_ms)[0];
console.log(fastest.name, fastest.response_ms + "ms");

Load into pandas (Python):

import pandas as pd
df = pd.read_csv("https://raw.githubusercontent.com/RelayStack/smmscout-data/main/data/panels.csv")
# full JSON payload (flags, liveness, per-panel inputs): https://smmscout.com/api/panels.json
print(df.sort_values("services", ascending=False).head(5))

Full interactive table: SMM panel data explorer (GitHub Pages) or the live directory.

This dataset is published by the team behind two SMM panels. They are scored by the identical formula and their numbers are not adjustable:

Both panels carry "owned": true in the data, and the site documents that no mechanism exists to adjust their numbers — see the disclosure. A recent catalog measurement dropped Smmize's score, and the drop was published in the changelog. That is the point of the dataset.

Score = longevity 30 (log-scaled, capped at 10 years) + catalog 25 (log-scaled, capped at 9,000 services) + API performance 35 + platform breadth 10 (capped at 8 platforms) + owner verification 25. Max 125. Every score in this dataset is recomputable from the published inputs. Full methodology: https://smmscout.com/methodology/

A panel carries a score only when its data is complete and a measurement says it is up: data_quality is not "pending" and liveness.state is "ok", which is exactly what the operational field reports. A dead or degraded domain is listed with "score": null — a number is never carried by a panel that is not answering.

Every panel also carries a risk score (risk_score, integer 0-100, higher = more caution), built from five other public inputs: active flags, domain age, a missing public price floor, pending data and owner verification. Bands: 0-19 low risk · 20-39 caution · 40-59 elevated · 60+ high risk. The https://smmscout.com/api/v2/panels.json endpoint publishes the matching risk_label. The risk score is a warning band, not the ranking — the Scout Score above is the ranking.

Is the SMM panel data free? Yes. JSON and CSV, no key, CORS open. The only ask: attribute SMMScout with a link when you republish.

How fresh is the data? Refreshed automatically every Monday via GitHub Actions; the generated_at field in panels.json records the pull time.

Why are some panels missing scores? Panels under review are listed with "data_quality": "pending" and no score, and so is a panel whose liveness probe did not find a healthy page ("operational": false, liveness.state not "ok"). Scores are never invented for them.

Can I use the data commercially? Yes, under CC BY 4.0 with attribution. The published formula may not be repackaged as a proprietary ranking.

Where does the data come from? Public measurements: WHOIS age, live catalog counts and API probes from a single vantage point, dated on every panel page.

Weekly via .github/workflows/refresh.yml (Mondays 03:00 UTC). Manual trigger: Actions → Refresh data → Run workflow.

Every push and pull request is checked by .github/workflows/verify.yml, so a dataset committed by hand cannot ship inconsistent with its derived files: the published formula is recomputed for every row, data/stats.json must equal what scripts/stats.mjs derives from the committed data/panels.json, and meta.json, data/panels.csv and the published panel counts must all agree. The job needs nothing but node — no npm install.

CC BY 4.0 — see LICENSE.

• Live directory: https://smmscout.com/panels/
• Methodology: https://smmscout.com/methodology/
• Market report: https://smmscout.com/market-report/
• Data API docs: https://smmscout.com/api/
• Scam reports & risk flags: https://smmscout.com/scam-reports/