EQL SourceSurvey

EQL Source  /  Public data

Take the data.
Please.

Nobody in this community publishes machine-readable data, so every tool re-transcribes the same wiki pages and inherits the same 1999 errors doing it. Here is ours: 4 datasets, versioned, with a source recorded per claim. Free to use, no key, no rate limit, CORS open.

4 datasets 313 KB total no key required

curl https://eqlsource.com/data/index.json

What is here

Plane of Sky class unlock quests

https://eqlsource.com/data/sky.v1.json

v1.0.0 · 125 KB · hash c8d41d080e92f873

Every Plane of Sky test for all sixteen classes, with the turn-ins, the island each component drops on, and a source recorded per claim.

Drops we have watched happen

https://eqlsource.com/data/sightings.v1.json

v1.0.0 · 163 KB · hash 8d93a88718ef7283

Every item we have personally seen drop, the mob it came from, and how many times. Parsed from our own combat logs.

Surveyed dungeons

https://eqlsource.com/data/zones.v1.json

v1.0.0 · 15 KB · hash 093bc71ba178315c

The zones we have surveyed, with the level band, the experience modifier and how far verification has got.

Item names and game IDs

https://eqlsource.com/data/items.v1.json

v1.0.0 · 9 KB · hash 519cde8b25db1101

Item name to the game's own numeric item ID, read from `/outputfile inventory` dumps.

What we promise

  • Fields in a vN file are never removed and never change type.
  • New optional fields may be added; ignore keys you do not recognise.
  • A breaking change gets a new version at a new URL. The old one stays up.
  • Values change as evidence improves. That is the product working, not a breach — watch `hash` if you cache.

What you may do with it

Use it, credit EQL Source, and read the provenance before you trust a value.

The measured data is ours. The measured data — drop sightings, damage figures, spell lists — is our own parsed play. Freely usable with attribution.

The inherited data is not. Fields sourced from community wikis carry their tier and source. Those values are not ours to license; respect the original source and do not present a tier 5 figure as fact.

No warranty. None. This is a fan project and the game is actively patched. Every dataset here states what it does not know.

What this data is not good for

Drop rates. The sightings file is a count of what we watched happen, and a drop seen once is seen once. There is no denominator in it and you cannot compute one. If you publish a percentage from this, you have invented it.

Anything marked below tier 2. Every claim carries its tier. A tier 5 figure is inherited Project 1999 prose that nobody has checked against Legends, and roughly a third of what any reference in this space publishes is that. We scan ourselves for it and publish what turns up. Read the provenance before you present a number as fact.

Completeness. Thirteen dungeons of far more, one plane of three surveyed in depth, and a sample of a few dozen play sessions. Everything here states what it does not know; that is the most useful field in the file.

If you build something with it

Tell us and we will link it. If a field you need is missing, or a shape is awkward to consume, open an issue — the point of publishing this is that it gets used, and a schema nobody can use is a press release.

And if you find something wrong in it, that is the most useful thing anyone can send us. Every finding is credited by name on the credits page.