Comprehensive judgment corpus
Income Tax and GST coverage — Supreme Court, High Courts, ITAT, GSTAT, AAR, plus statutory text and CBDT/CBIC circulars. Every hit comes back with citation, court level, and outcome where available.
Indian tax case judgments via production REST API. Income Tax + GST. Drop into your stack and ship legal-research features in days.
Public, fixed pricing. Annual prepayment earns a 10% discount automatically.
per month, billed annually
per month, billed annually
per month, billed annually
per quote
Effective per-month price shown — billed annually upfront, 10% discount applied. GST 18% extra. Late payments: 1.5%/month interest. Onboarded within 48 hours of contract sign-off.
REST API. JSON in, JSON out. Standard X-API-Key header. Real responses in 200-500ms.
curl -X POST https://api.taxnoticeai.app/api/v1/caselaw/search \
-H "X-API-Key: tnk_live_demo_replace_with_your_key" \
-H "Content-Type: application/json" \
-d '{
"query": "deemed dividend section 2(22)(e) loan to shareholder",
"jurisdiction": "income_tax",
"court_level": "itat",
"top_k": 10
}'Demo prefix only — your live key is handed over during onboarding.
Income Tax and GST coverage — Supreme Court, High Courts, ITAT, GSTAT, AAR, plus statutory text and CBDT/CBIC circulars. Every hit comes back with citation, court level, and outcome where available.
Most relevant cases first. Filter by section, court level, jurisdiction, date range, or outcome — the API does the heavy lifting so your product can show the right answer in milliseconds.
Every result links to its citation graph and statutory references. Outcome inferred where the judgment supports it. No post-processing required.
The corpus is regularly refreshed — new judgments are ingested, citations re-linked. You always query the latest version. No stale indices.
The kind of citation graph legacy legal databases charge ₹50K-1L per seat per year for. Bundled with our Pro tier at ₹12L/month flat — unlimited users, unlimited downstream customers. Same-day access.
Followed, distinguished, overruled, referred, explained, approved, disapproved — for every citation, with the supporting paragraph extract.
See which cases a judgment cites, AND which later judgments cite it back. Filter by treatment + court level.
/citation-graph?depth=2 returns the full neighbourhood as nodes + edges JSON. Built-in visualization in the search UI.
Submit a free-text fact pattern. Get back a calibrated assessee win-probability, the precedents driving the prediction, and a factor breakdown by court tier. Production-ready REST endpoint. Pro tier and above.
Aggregated from the most similar judgments, weighted by court tier and recency. Returns a calibrated 0-100% probability with a separate confidence score.
Every prediction ships the top 10 most similar precedents inline — citations, similarity scores, outcome labels — so your team can audit the call.
Signed contributions per court tier (Supreme Court, High Court, ITAT, GSTAT, AAR) plus the section-level historical win-rate baseline when a section is supplied.
Pass as_of_date=YYYY-MM-DD on any caselaw query and get back only the judgments that existed (and were not yet overruled) on that date. Forensic accountants, audit defense, litigation strategy: reconstruct the legal landscape your client was working with.
“At the time of filing, the law as understood was…” — back the position with the precedents available on that date.
Cite only the cases that existed at the time of the assessee’s filing. The corpus filter does it; you don’t have to vet every result manually.
Judgments overruled before your as-of date drop out of the result set. The citator (Pro+) feeds the time-machine (Business+).
Onboarded within 48 hours of contract sign-off. NEFT/RTGS billing — no payment processor in the loop.