Vedic Jyotish computation engine + consumer guidance platform.
1338 tests | Sessions 1–188 complete | 120+ modules | Engine v3.0.0
Classical Audit — March 2026: An independent audit against BPHS, Phaladeepika,
Saravali, Brihat Jataka and Jaimini Sutras is complete. The scoring engine is a
heuristic — additive numeric weights are not a classical methodology. All Phase 0
correctness fixes are complete (Sessions 101–188). Swiss Ephemeris (JPL DE431)
active — Moshier fallback retired. See AUDIT.mdPLAN.md
A structured docs/ folder contains all development documentation:
docs/
├── MEMORY.md Session state — read this first each session
├── ARCHITECTURE.md Module APIs, function signatures, tacit knowledge
├── BUGS.md Bug status and active issues
├── CHANGELOG.md Full session history
├── GUARDRAILS.md 24 guardrails G01–G24
├── CLASSICAL_CORPUS.md Text inventory and encoding plan
├── KPIS.md KPI scorecard
├── RESEARCH.md Calibration, OSF, SHAP framework
├── ROADMAP.md Phase plan S189–S1050+
├── PREDICTION_PIPELINE.md 10-layer prediction architecture
└── SESSION_TEMPLATE.md Reusable session template
A comprehensive Jyotish platform covering:
Natal analysis — 63 calculation modules, 23 BPHS rules × 12 houses × 5 lagna axesMuhurta — electional astrology with Panchanga, Tarabala, Chandrabala, Hora, ChoghadiyaPrashna — horary Jyotish from query-moment chartAll major dashas — Vimshottari, Narayana, Yogini, Chara, Kalachakra, Ashtottari, Shoola, Sudasa, Tara, Drig, Lagna KendradiUpaya — classical remedial measures (gemstones, deities, mantras, charity)Mundane — nation charts, solar/lunar ingress, swearing-in chartsConsumer layer — Bloomberg-style guidance with safety pipeline, GDPR compliance, feedback governance
git clone https://github.com/agniinvestor/LagnaMaster.git
cd LagnaMaster && docker compose up --build Key API Endpoints (v3.0.0)
Endpoint
Description
POST /chartsCompute + store a birth chart
GET /charts/{id}/scores22-rule house scores
GET /charts/{id}/scores/v3Dasha-sensitized multi-axis scores (D1/D9/D10/CL/SL)
POST /charts/{id}/svgNorth/South Indian chart SVG
POST /charts/{id}/pdf2-page PDF export
POST /charts/{id}/guidanceConsumer guidance (L1/L2/L3)
GET /charts/{id}/confidenceLagna/nakshatra boundary confidence
Key API Endpoints (v3.0.0)
Endpoint
Description
POST /chartsCompute + store a birth chart
GET /charts/{id}/scores22-rule house scores
GET /charts/{id}/scores/v3Dasha-sensitized multi-axis scores (D1/D9/D10/CL/SL)
POST /charts/{id}/svgNorth/South Indian chart SVG
POST /charts/{id}/pdf2-page PDF export
POST /charts/{id}/guidanceConsumer guidance (L1/L2/L3)
GET /charts/{id}/confidenceLagna/nakshatra boundary confidence
python3 -m venv .venv && source .venv/bin/activate
pip install -r requirements.txt
ulimit -n 4096
PYTHONPATH=. pytest tests/ -q -p no:warnings Muhurta (Electional Astrology) from src .calculations .panchanga import compute_panchanga , compute_hora , compute_choghadiya
from src .calculations .muhurta import score_muhurta
from datetime import datetime
p = compute_panchanga (sun_lon , moon_lon , datetime .now ())
score = score_muhurta ("marriage" , p ,
birth_nakshatra = 7 ,
birth_moon_sign = 3 ,
muhurta_lagna_sign = 1 )
print (score .quality ) # "Excellent"/"Good"/"Acceptable"/"Avoid"
print (score .total_score ) # 0–7
print (score .warnings )from src .calculations .prashna import analyze_prashna
result = analyze_prashna (prashna_chart , query_type = "career" ,
query_dt = datetime .now ())
print (result .verdict ) # "Yes — strongly indicated" / "Unlikely" etc.
print (result .confidence ) # "High"/"Moderate"/"Low"
print (result .reasoning )from src .calculations .vimshottari_dasa import compute_vimshottari_dasa # 120yr, 9 planets
from src .calculations .narayana_dasha import compute_narayana_dasha # 81yr Rasi dasha
from src .calculations .yogini_dasha import compute_yogini_dasha # 36yr cycle
from src .calculations .chara_dasha import compute_chara_dasha # Jaimini
from src .calculations .kalachakra_dasha import compute_kalachakra_dasha # 100yr, Moon D9
from src .calculations .ashtottari_dasha import compute_ashtottari_dasha # 108yr (conditional)
from src .calculations .shoola_dasha import compute_shoola_dasha # Longevity/Ayur
from src .calculations .shoola_dasha import compute_sudasa # Material success
from src .calculations .tara_dasha import compute_tara_dasha # 9-category nakshatra Upaya (Remedial Measures) from src .calculations .upaya import get_chart_upayas
upayas = get_chart_upayas (chart )
for u in upayas :
print (f"{ u .planet } { u .affliction_type } )
print (f" Gemstone: { u .gemstone } { u .gemstone_metal } { u .gemstone_day } )
print (f" Deity: { u .primary_deity } )
print (f" { u .disclaimer } ) # always present — never removed Phase 19 — Advanced Dashas, Yogas & Plugin Architecture # Drig Dasha (Session 101) — PVRNR preface p8
from src .calculations .drig_dasha import compute_drig_dasha
periods = compute_drig_dasha (chart , birth_date ) # 12 sign-based periods
# Lagna Kendradi Dasha (Session 102)
from src .calculations .lagna_kendradi_dasha import compute_lagna_kendradi_dasha
periods = compute_lagna_kendradi_dasha (chart , birth_date ) # Kendra→Panapara→Apoklima
# K.N. Rao Double Transit (Session 103)
from src .calculations .double_transit import compute_double_transit
r = compute_double_transit (chart , transit_date , domain = "marriage" )
print (r .signal ) # "Double confirmation — event likely"
# Upapada Lagna (Session 104) — Arudha of H12
from src .calculations .upapada_lagna import compute_upapada
ul = compute_upapada (chart )
print (ul .marriage_quality ) # "Favourable"/"Mixed"/"Challenging"
print (ul .marriage_longevity ) # "Lasting"/"Possible challenges"/"Separation risk"
# Kala Sarpa Yoga (Session 105) — modern practitioner convention
from src .calculations .kala_sarpa import compute_kala_sarpa
ks = compute_kala_sarpa (chart )
print (ks .present , ks .yoga_name , ks .is_reverse ) # Kala Amrita if reverse
print (ks .classical_disclaimer ) # always present — not in BPHS
# Nabhasa Yogas (Session 106) — 32 types from BPHS Ch.35
from src .calculations .nabhasa_yogas import detect_nabhasa_yogas
yogas = [y for y in detect_nabhasa_yogas (chart ) if y .present ]
for y in yogas :
print (f"{ y .name } { y .group } { y .result } )
# Pitr Dosha (Session 107) — modern practitioner convention
from src .calculations .pitr_dosha import compute_pitr_dosha
pd = compute_pitr_dosha (chart )
print (pd .severity ) # "Strong"/"Moderate"/"Mild"/"Not present"
print (pd .classical_disclaimer ) # always present — not in BPHS
# Rule Plugin Architecture (Session 108)
from src .calculations .rule_plugin import register_yoga , apply_all_plugins
@register_yoga ("My Custom Yoga" , source = "Regional tradition" , score_if_present = 1.5 )
def my_yoga (chart ) -> bool :
return chart .planets ["Jupiter" ].sign_index == chart .lagna_sign_index
results = apply_all_plugins (chart )from src .guidance .guidance_api import get_guidance
r = get_guidance (chart , domain = "career" , depth = "L1" , dashas = dashas )
print (r .signal_display ) # ●●●○○
print (r .timing_label ) # "Mixed — lean in"
print (r .summary ) # Human-safe sentence; no raw scores at L1/L2
print (r .disclaimer )
Phase
Sessions
Status
Key Deliverables
1–3
1–27
✅
Engine, scoring, FastAPI, Streamlit, Docker, JWT, K8s
4–5
28–40
✅
LPI, 5-axis scoring, rule interactions, Scoring v3
6
41–48
✅
Ishta/Kashta, Longevity, Yogini, KP, 200+ Yogas, Empirica
7
49–56
✅
Sayanadi (12-state), Panchadha Maitri, Dig Bala continuous, Config toggles
8
57–63
✅
Orb strength, Yoga fructification, Stronger-of-two, AV transit
9
64–70
✅
Dominance engine, Promise/Manifestation, Confidence model
10–11
71–78
✅
Language safety pipeline, GDPR/DPDP consent + erasure
12–14
79–90
✅
Bloomberg UI, feedback governance, educational layer, mobile router
15–18
91–100
✅
Muhurta, Prashna, Kalachakra, Ashtottari, Shoola, Tara, Upaya, Mundane
19+
101–108
🔄
Phase 0 classical correctness fixes (on remote — git pull to get)
Branch
Status
Natal analysis (BPHS + PVRNR)
✅ Complete
Divisional charts (20 vargas)
✅ Complete
All 9 dasha systems
✅ Complete
Muhurta (electional)
✅ Session 92
Prashna (horary)
✅ Session 93
Upaya (remedial measures)
✅ Session 97
Mundane astrology
✅ Session 98
Consumer safety + GDPR
✅ Sessions 71–78
Bloomberg UI (Next.js)
✅ Built — integration testing pending
Phase 0 correctness (classical audit)
🔄 Sessions 101+
Acknowledged limits (correctly excluded):
Full Prashna Marga corpus · Desha-Kala-Patra in full · Gestalt expert synthesis · Medical/financial astrology
Layer
Technology
Ephemeris
pyswisseph (Lahiri / Raman / KP / Fagan-Bradley)
Backend
FastAPI + Celery + Redis
Database
SQLite → PostgreSQL + consent/feedback tables
Consumer UI
Next.js 14 + TypeScript + Tailwind
Mobile
FastAPI /mobile router (React Native shell pending)
Deploy
Docker Compose + Kubernetes Helm
Auth
JWT + bcrypt
CI/CD
GitHub Actions
Privacy
GDPR Art.7+17 · DPDP · CCPA/CPRA
Document
Purpose
PLAN.mdFull session-by-session build plan and phase roadmap
docs/MEMORY.mdProject state for Claude Code sessions — module inventory, invariants
AUDIT.mdClassical audit findings — gaps, correctness issues, cited ślokas
CHANGELOG.mdSession-by-session delivery history