Crypto Fear and Greed Index Scraper

Crypto Fear and Greed Index Scraper

The Crypto Fear and Greed Index is one number a day between 0 and 100, with a word attached to it: Extreme Fear, Fear, Neutral, Greed, Extreme Greed. This actor hands you that number as dated rows: today's reading, or every reading since 1 February 2018, or any window in between.

One number a day is a thin thing to buy on its own, so two other free daily indexes ride along in the same row shape. You pick which ones you want.

Where the numbers come from

Nothing here is estimated, modelled or reconstructed. Every value is the publisher's own, read from the publisher's own feed.

The Fear & Greed Index data is provided by alternative.me. Their terms allow commercial use with attribution, so every row carries source and sourceUrl, and if you republish the numbers you need to credit them where the numbers appear.

The two Fear & Greed indexes disagree, often by ten points or more, because they weight different inputs. That is the reason to pull both rather than a defect in either.

Input

The smallest useful run:

{ "sources": ["fear-greed"], "historyDays": 1 }

That returns today's reading. One row.

Thirty days of both Fear & Greed readings, so you can line them up:

{ "sources": ["fear-greed", "cmc-fear-greed"], "historyDays": 30, "maxItems": 60 }

A specific stretch of history, for a backtest:

{
  "sources": ["fear-greed"],
  "dateFrom": "2024-01-01",
  "dateTo": "2024-12-31",
  "maxItems": 400
}

Everything, every index, since each one began, about 5,200 rows:

{ "sources": ["fear-greed", "cmc-fear-greed", "altcoin-season"], "historyDays": 4000, "maxItems": 10000 }

maxItems is split evenly between the indexes you picked, newest first. Ask for two indexes and 20 rows and you get ten days of each, not twenty days of whichever was read first. The run logs how many readings exist in your window before it returns anything, so you can see what you left behind and raise the cap if you want it.

Output

One row per index per day. A real row, from a run on 2026-09-20:

{
  "ok": true,
  "date": "2026-09-20",
  "timestamp": "2026-09-20T00:00:00.000Z",
  "index": "fear-greed",
  "indexName": "Crypto Fear & Greed Index",
  "value": 71,
  "classification": "Greed",
  "classifiedBy": "publisher",
  "scaleMin": 0,
  "scaleMax": 100,
  "isLatest": true,
  "source": "alternative.me",
  "sourceUrl": "https://alternative.me/crypto/fear-and-greed-index/",
  "btcPriceUsd": null,
  "btcVolume24hUsd": null,
  "altcoinMarketCapUsd": null,
  "scrapedAt": "2026-09-20T00:23:15.336Z"
}

And a CoinMarketCap Fear & Greed row from the same run, which fills in the price fields:

{
  "ok": true,
  "date": "2026-09-20",
  "index": "cmc-fear-greed",
  "indexName": "CoinMarketCap Fear & Greed Index",
  "value": 73,
  "classification": "Greed",
  "classifiedBy": "publisher",
  "btcPriceUsd": 81179.95,
  "btcVolume24hUsd": 20388570058.78,
  "source": "coinmarketcap.com"
}

classifiedBy is there because one of the three indexes does not ship a label with every data point. Both Fear & Greed feeds do, and those rows say publisher. The Altcoin Season feed leaves the label blank on historical points, so this actor applies the publisher's own stated cut-offs of 75 and above, 25 and below. Those rows are marked publisher thresholds, so you know the value is theirs and the word is not.

btcPriceUsd, btcVolume24hUsd and altcoinMarketCapUsd are null on the indexes that do not publish them. They are never filled in from somewhere else.

isLatest is true on the newest row of each index in that run.

What it costs

You are charged per reading returned. Nothing else bills.

• The sample row you get from an empty sources list is free.
• Diagnostic rows are free. If an index cannot be reached, or your window predates it, you get a row saying so and you are not charged for it.
• A run that returns no readings at all charges nothing beyond the run fee.
• maxItems is honoured before anything is fetched, so the row count you set is the row count you pay for.

The live per-row figure is on the Pricing tab of this page. It is a flat rate with no volume tiers, so a thousand rows cost a thousand times one row and there is nothing to calculate.

What this does not do

Read this part before you buy.

It is not a trading signal and it is not advice. A sentiment index is a description of the mood a market was in on a given day. It has no predictive claim attached to it by its publishers and none is made here. If you are looking for something that tells you when to buy, this is the wrong purchase however cheap it is.

There is no per-coin Fear & Greed here. Per-coin sentiment scores do exist, but the publishers that produce them put them behind an account and an API key. This actor uses no accounts and no keys, so it ships what is genuinely open and says nothing about the rest.

Daily only. These indexes are published once a day. There is no hourly or 15-minute series in any of these feeds, so there is none in the output.

The two Fear & Greed publishers capitalise their labels differently. alternative.me writes Extreme Fear; CoinMarketCap writes Extreme fear. Rows carry whatever the publisher wrote, unchanged. Lowercase the field before you group by it across both indexes.

Coverage starts where each publisher's history starts, not where you ask. A window in 2019 returns rows from the alternative.me index and an uncharged diagnostic row from the other two, because they did not exist yet.

Today's reading moves during the day. All three indexes update as the day goes on, so a row dated today is a live value, not a settled close. Rows for past days are final.

Four days are missing from the alternative.me history. 14, 15 and 16 April 2018 are absent, and so is 26 October 2024. Those are gaps in the publisher's own series. They are passed through as gaps rather than filled in, so a 3,150-row full pull covers 3,154 calendar days.

FAQ

What is the Crypto Fear and Greed Index? A 0-100 score published daily by alternative.me that summarises how fearful or greedy the crypto market looks that day. 0 is Extreme Fear, 100 is Extreme Greed. It leans on Bitcoin volatility, trading volume, social media activity, Bitcoin dominance and search trends.

How far back does the history go? To 1 February 2018 for the alternative.me index, which is 3,150 daily readings as of September 2026. The CoinMarketCap score starts 29 June 2023 and Altcoin Season starts 22 March 2024.

Do I need an API key or an account? No. Not for this actor and not for the feeds behind it.

Can I get just today's number? Yes. {"sources": ["fear-greed"], "historyDays": 1} returns one row.

How do I get the whole history without a surprise bill? maxItems caps it. Set it to what you are willing to pay for and the run stops there, newest readings first. The full history of all three indexes is about 5,200 rows, so leaving maxItems at its default of 400 is what stops a "give me everything" run from being larger than you meant.

Why are there two Fear & Greed indexes? Because two organisations publish one and they do not agree. Pulling both gives you a spread, and the CoinMarketCap rows carry the day's Bitcoin price so you can put the score next to what the market actually did.

What is the Altcoin Season Index? The share of the top 100 coins, excluding stablecoins and wrapped tokens, that outperformed Bitcoin over the previous 90 days, scaled to 1-100. At 75 or above CoinMarketCap calls it Altcoin Season; at 25 or below, Bitcoin Season.

What happens if a feed is down? You get an uncharged diagnostic row naming the index and the error, and the other indexes you selected still return their rows. The run ends SUCCEEDED rather than failing, so a half-outage does not cost you a failed run.

Can I run this on a schedule? Yes, and daily is the sensible cadence given the data updates once a day. {"sources": ["fear-greed"], "historyDays": 1} on a daily schedule appends one row a day.

Is the data licensed for commercial use? alternative.me allows commercial use of their index with attribution shown next to the data. Every row carries source and sourceUrl so you have what you need to credit them. Check the publishers' own terms for your specific use.