Calibration · archived run

How well does pkgxray call it?

An at-scale, code-only static scan of published packages, measured against a committed known-malware corpus. Four numbers, adjudicated by hand, reproducible from the inputs below.

1,301
packages scanned
published npm + MCP packages, one static pass
0.1%
false blocks, top 1,000
1 of the 1,000 most-downloaded packages
90.0%
known-malware catch rate
18 of 20 corpus samples blocked outright
2026-07-19
run date
pkgxray 1.0.3 · build 250a3b3

Aggregate calibration of a one-time at-scale static scan. Catch rate is measured against the committed reconstructed known-malware corpus (npm removes live malware, so live recall is untestable). The one top-1000 false block is the as-measured figure that drove the retune; the full scan was not re-run on the fixed engine. See methodology.

Corrections

Published runs are immutable. If a number is wrong we publish a new dated run and note it here — we never silently edit a number in place. Contest a figure at the tracker.

  • 2026-07-19 — Full 1,000-package scan re-run on the retuned engine; the single pre-retune heuristic false block now resolves to review, with no new false block. (superseded by 2026-07-19-retuned)

Run history

Every published run stays up at a stable URL.

Pinned version: this page is the 2026-07-19 snapshot at pkgxray 1.0.3 · 250a3b3. Latest run →