MaintainerPulse

temporal prediction of dependency maintenance risk

anthropic anthropics/anthropic-sdk-python ✓ low risk

no release in the next 12 months

28%

new issues will go unanswered (30 days)

39%

maintainer activity collapse within 12 months

100%

chance of a release within 12 months (survival model)

Model risk over time

calibrated P(no release next 12mo) at each historical monthly snapshot

Repository activity, last 36 months

pushes issues PyPI release

What drives this score

TreeSHAP contributions; raises risk / lowers risk

Survival curve

P(still no release) m months ahead, discrete-time hazard model

What if? — poke the model

drag a signal and the deployed models rescore this package live (implied signals move together — two years without a release also zeroes "releases, last 12mo"). The needle tracks the activity-collapse model, a logistic regression: monotone by construction, so it responds smoothly to counterfactuals where the tree models step.

maintainer activity collapse, 12mo

In plain words

only 4% of recent issues get a response in 30 days
healthy cadence: 91 releases in the last year
active team: 12 people pushed in the last year

Signals at the latest snapshot

days since last release	9	releases, last 12mo	91
people pushing, last 12mo	12	bus factor (top pusher share)	40%
issues opened, last 12mo	24	30-day response rate	4%
stars accumulated	2397	open vulns without fix	0

Probabilities are isotonic-calibrated on a held-out validation year and evaluated on future snapshots the models never saw (PR-AUC 0.894, precision@50 = 1.00 on the headline target).