Skip to main content
Engraving of a Roman archive hall with shelves of scrolls
GET https://palatineai.site/projects.json
→ 200 · application/json · ~140 KB · no auth · ETag + Last-Modified
One JSON object, two top-level keys:
KeyTypeContents
statsobjectDatabase-wide counts and the dataset date
projectsarrayThe top 120 scored projects, sorted by rank ascending

​
stats

FieldTypeCurrent valueMeaning
totalProjectsint2857Projects researched in the full database
scoredProjectsint1680Fully scored across all 12 signals
manualReviewint1177Still queued for manual scoring — full DB only, never published
categoriesint33Categories in the full database (22 appear in the published feed)
countriesint108Countries in the full database (35 in the feed)
noTokenSignalsint1599Scored projects with no token signal
topProjectstring"QSTN"Name of rank 1
publishedProjectsint120Length of projects
updatedstring"2026-06-11"Dataset version stamp, YYYY-MM-DD

​
Project object

FieldTypeNotes
projectstringProject name. Unique within the feed.
descriptionstringTeam-submitted summary, verbatim. Length varies widely.
categorystringOne of 22 exact strings — list in the cookbook.
countrystringOrigin country. Occasionally informal (e.g. "Ireland {Republic}") — normalize before grouping.
teamstringFounder name(s) and handle(s), free-form.
websitestringOfficial site. Empty string for 4 of 120 projects — fall back to x.
xstringX (Twitter) profile URL.
scoresobject12 signals, each 0–10 float — see The 12 signals.
totalfloatSum of the 12 signals. Max 120. Feed range today: 74.75–83.75.
rankintPosition by total among all 1,680 scored projects. Contiguous 1–120 in the feed.
tokenStatusstringExactly "No token signal" or "Possible token/funding signal".
manualReviewboolAlways false in the published feed (unscored projects are never published).
sheet, sheetRankstring, intInternal research-desk bookkeeping. Ignore.

​
Example record

Rank 1 as currently published (description abridged here, full in the feed): 
{
  "project": "QSTN",
  "description": "QSTN V2 is a Web3 data-monetization platform that turns everyday social engagement into user-owned income. …",
  "category": "Consumer Apps",
  "country": "United States",
  "website": "https://data-earn.qstn.us",
  "team": "QSTN (@qstn)",
  "x": "https://x.com/qstnus",
  "scores": {
    "githubActivity": 4.25, "githubQuality": 3.5, "branding": 8.0,
    "country": 10.0, "teamExperience": 5.0, "ideaUniqueness": 10.0,
    "ideaQuality": 6.5, "idea": 9.0, "socialActivity": 7.5,
    "noToken": 10.0, "funding": 10.0, "additional": 0.0
  },
  "total": 83.75,
  "tokenStatus": "No token signal",
  "sheet": "Projects",
  "manualReview": false,
  "rank": 1,
  "sheetRank": 1528
}

​
Parsing guarantees

  • projects is pre-sorted by rank; projects[0] is always the top project.
  • All 12 score keys are present on every record; no nulls in the feed.
  • Strings may be empty (website) but keys are never missing.
  • The schema is additive: new fields may appear without notice. Ignore unknown keys; don’t validate strictly against this page.