AstroGuru

Research-grade Vedic astrology, validated against data.

A computational platform for serious astrology study and statistical validation of classical principles. Built for students, researchers, and practicing astrologers who want evidence alongside tradition.

What we offer

Free

Free Chart Calculator

14 divisional charts (D1 through D60 plus karakamsha)
Vimshottari, Yogini, Chara and Kaala Chakra dashas with interactive timelines
Yogas, Shadbala, Ashtakavarga, KP sublord, Jaimini karakas
North Indian and South Indian chart styles
Comprehensive JSON download for further analysis
Print-friendly layout

Open Calculator

$19/mo or $190/yr

Research Platform

Test hypotheses against 76,133 computed charts
104,904 life events spanning birth to death
Natural-language queries translated to validation predicates (Phase L2)
Wilson 95% confidence intervals on every result
Beta-Binomial Bayesian analysis for small subsets
Public hypothesis library plus private research
Free tier: read-only access to public hypotheses

Open Research Platform

Included

API + MCP Access

REST API: chart compute, hypothesis validation, corpus stats
MCP server for Claude Desktop integration
5 tools: compute_chart, run_hypothesis, get_validation_run, list_public_hypotheses, get_panchanga
Programmatic access to the entire platform
Included with Research Platform subscription

View Documentation

How validation works

The corpus

79,654 birth-data records sourced from AstroDB (Astrodatabank), the largest open corpus of charts with verified birth times and biographical event data.

76,133 fully computed charts (95.6% success rate) using Swiss Ephemeris with Lahiri ayanamsa
104,904 life events: births, marriages, accidents, deaths, career milestones, recognitions
15 divisional varga charts per record: D1, D2, D3, D7, D9, D10, D12, D16, D20, D24, D27, D30, D40, D60, plus karakamsha

The math

Every reported result carries an honest uncertainty estimate.

Precision and recall Precision = TP / (TP + FP) Recall = TP / (TP + FN)

Wilson 95% confidence interval on a proportion p_w = ( p + z²/(2n) ± z·√( p(1−p)/n + z²/(4n²) ) ) / ( 1 + z²/n )
with z = 1.96 for 95% confidence.

Wilson handles small samples and proportions near 0 or 1 correctly, unlike the Gaussian approximation.

Beta-Binomial posterior for small subsets (n < 100) Posterior ~ Beta(α + TP, β + FP)
with uniform Beta(1, 1) prior. Report posterior mean and 95% credible interval.

Bonferroni correction for multiple-hypothesis testing α_adjusted = α / m
where m is the number of hypotheses tested.

Classical systems supported

Parashara (Brihat Parashara Hora Shastra) — primary
Krishnamurti Paddhati (KP) with sublords and significators
Jaimini (with chara karakas, karakamsha)
Yogini dasha (positive and negative signs)
Nadi astrology cross-references
Western Tropical cross-reference (where useful)

Full mathematical treatment, predicate vocabulary reference, and corpus methodology at docs.astroguru.cc.

Sources and privacy

AstroDB attribution: Birth chart and life event data sourced from AstroDB (Astrodatabank), maintained by Astrodienst AG and the community of contributing astrologers. AstroGuru is a research platform built on this corpus and is not affiliated with Astrodienst.
Aggregate-first: All statistical results display as aggregate counts and rates by default. Individual chart drilldown requires explicit per-chart confirmation.
Death-event policy: Death-timing analysis is available for research purposes only. AstroGuru never produces consumer-facing death predictions.
Privacy: User-submitted charts are private by default. Hypothesis submissions are private unless explicitly published to the public library.