Claw IP and Court Data API: Endpoints, Rate Limits, Pricing and Coverage

Developer Guide · IP and Court Data API

Updated for 2026 · Claw Editorial

A developer who needs Indian IP and court data usually finds two disconnected worlds: registry data from IP India on one side, and case data spread across hundreds of court websites on the other. Neither side offers one clean, documented, rate-limited API you can build a product on. This guide explains what to look for in an IP and court data API for India, which numbers to ask every vendor for, and how to plan an integration that will not break in production. Last verified October 2026.

The short answer

Two data sources: IP registry records (IP India) and court records. A good API joins both.
Ask every vendor for five numbers: requests per minute, requests per month, price per 1,000 calls, p95 latency, and the share of TM records with a non-null status.
Test coverage yourself: use a sample of 100 to 200 known matters, and record N and the test date.
Claw: offers an API, covers 9,102 courts and 30 crore+ judgments. Get the live spec for exact limits.
Last verified: October 2026.

01Why IP and court data is hard to get through an API

Indian IP and court data is public in principle, but it is not built for programmatic use. A developer runs into the same three walls again and again.

The data lives in two systems

Trademark, patent, design and GI records sit with IP India. Litigation about those rights sits in court systems: the Supreme Court, the High Courts (including the IP Divisions of some High Courts), district courts and tribunals. A product that needs to answer "is this mark registered, opposed or being litigated?" must join both sides.

Court websites are not uniform

Every court publishes cases in its own format. Case numbers, party names and case-type codes differ from court to court. Party names are spelt in many ways. Matching one entity across courts is a data engineering problem, not just an HTTP call.

Scraping breaks and has legal risk

Many teams start by scraping. It works for a demo and fails in production, because pages change, captchas appear and rate blocks arrive without warning. A licensed API moves that maintenance burden to the vendor.

Two jobs, one integration

For the workflow view of joining registry and litigation data, see IP litigation and registry in one workflow. This page stays on the API side: what to ask for, and how to plan the build.

02What a good IP and court data API must offer

A usable API for this job has to be documented, predictable and honest about its limits. Check each of these before you write any code.

A public, versioned spec: an OpenAPI or Postman collection you can read before you sign up, so you can judge fit without a sales call.
Client libraries: a public GitHub repository with an SDK in the language you use, or at least working code samples.
Clear limits: requests per minute and per month, written down, with the error code you get when you cross them.
Stated latency: a p95 response time, not only a best-case number.
Stated coverage: which courts, which years and which IP registries, with a test date and sample size for any accuracy figure.
Transparent pricing: a price per 1,000 calls, or a plan with a clear call allowance, with GST status stated.

The five numbers every Indian IP and court data API should publish: requests per minute, requests per month, price per 1,000 calls, p95 latency, and the share of records with a non-null trademark status.

03Endpoints to look for

Whatever vendor you pick, an IP and court data API for India should map to a small set of resources. Use this list as a checklist when you read a spec.

Resource	What you should be able to do	Why it matters
Trademark records	Look up by application number or mark text; read status, class, owner and dates	Status fields drive alerts and renewals
Trade Marks Journal	Fetch journal entries and run text, phonetic or visual conflict checks	Watch services depend on it
Patent, design and GI records	Look up by application number; read status and key dates	Needed for annuity and response deadlines
Case search	Search by party name, case number, court and year	Finds infringement and opposition litigation
Case detail	Read hearings, orders and next date for one case	Feeds calendars and alerts
Cause lists	Read daily listings by court or bench	Shows what is listed tomorrow
Judgments	Search and fetch judgment text and citations	Research and precedent checks
Webhooks or polling	Get notified on a status or hearing change	Avoids heavy polling

Some High Courts handle IP cases through a dedicated IP Division with its own case-type codes. Confirm the code format against the court’s own rules before you hard-code it, since codes can change.

For registry data, remember the legal frame your data points to. A registered trademark in India runs for ten years from filing and can be renewed, and a patent runs for twenty years from the filing date. These terms come from the Trade Marks Act, 1999 and the Patents Act, 1970.

04Rate limits, latency and pricing: what to ask

The numbers below decide whether an API fits your product. Ask every vendor for them in writing, and test them yourself.

Rate limits

Ask for the limit per minute and the limit per month. A per-minute limit shapes how you design bulk jobs. A per-month limit shapes your cost. Also ask what happens at the limit: a clear error with a retry hint is workable, a silent block is not.

Latency

Average latency hides slow requests. Ask for the p95 figure and measure it from your own servers in India, on your own queries.

Pricing

Compare vendors on price per 1,000 calls and on GST status. Some vendors price by subscription, some by call volume, and many quote only after a demo. Never compare a per-seat subscription to a per-call price without converting both to the same unit for your expected volume.

We have not listed competitor API prices here, because most vendors publish no public API price. Where a vendor does publish a price, source it to that vendor’s own page, with the date and whether GST is included.

Public alternatives

For small, low-volume needs, free public sources exist. IP India offers public search for trademarks, patents and designs, and the eCourts portal offers public case status. Indian Kanoon is a widely used public judgment search. These are useful for checks and for verifying vendor data, but they do not offer one documented, rate-limited API that covers registry and courts together.

05How to measure coverage before you commit

Coverage is a number you should test, not a claim you should accept. Vendors say "all courts", and the gap between that phrase and your actual cases can be large.

Build a sample: pick 100 to 200 real matters you already know: a mix of trademark, patent and design cases, across courts and years.
Run them through the API: record how many are found, how many return complete party names, and how many registry records come back with a non-null status.
Write down N and the date: a result such as "found 92 of 100" is only meaningful with the sample size and the test date next to it.
Repeat monthly: coverage and data quality drift. A one-time test is not enough.

Pay special attention to the share of trademark records with a non-null status. A record with a blank status cannot drive an alert, so this single figure tells you how usable the registry data is for monitoring.

For monitoring many parties at once, see bulk litigation search and monitoring tools. For a worked example of a data report built from registry records, see the university patent lapse report.

06Plan a first integration

A safe first integration is small, observable and reversible. Use this order of work.

Read the spec first. Import the OpenAPI or Postman file, and call two or three endpoints by hand.
Store keys safely. Keep API keys in server-side secrets, never in browser code or a public repository.
Add retries with backoff. Treat rate-limit errors as normal, and queue bulk jobs below the per-minute limit.
Cache what rarely changes. A closed case or a granted registration changes slowly, so cache it and refresh on a schedule.
Normalise names. Party and applicant names vary. Keep both the raw string and your cleaned version.
Log every call. Record the endpoint, latency and status code so you can check p95 against what the vendor promised.
Keep a human check. For any legal deadline, such as a renewal or a hearing date, confirm against the official source before acting.

Also check how personal data is handled. Party names in case records are personal data, so plan for the Digital Personal Data Protection Act, 2023 in how you store and share them.

07Where Claw fits

Claw is an all-in-one legaltech platform for Indian advocates, law firms, and corporate legal teams, combining AI-based case search, an AI legal assistant (Legal GPT), case management, and compliance automation across all Indian courts and tribunals.

For the court side of this data, Claw gives you one licensed source instead of scraping. It holds 30 crore+ judgments and tracks cases, orders and hearings across 9,102 courts, benches and forums, with Supreme Court records from 1950 and High Court records from 1990, and it returns results in under 5 seconds. Claw does not use customer documents to train AI models, which matters when you plan how to store and share what you pull.

Create a FREE Claw account to test coverage on your own sample of matters before you build, and to run case search across Indian courts.

08Sources and further reading

Check any statute, fee or form against these primary sources before relying on it. Last verified October 2026.

IP India (trademark, patent, design and GI records and fees): ipindia.gov.in
eCourts Services (public case status): ecourts.gov.in
WIPO (international IP standards and data): wipo.int
Indian Kanoon (public judgment search): indiankanoon.org
Claw: clawlaw.in

This guide gives no competitor prices. Confirm every vendor’s limits, latency and pricing with the vendor directly.

09Frequently asked questions

Is there an IP and court data API for developers in India?

Yes. Claw offers an API for programmatic access to court and litigation data, and its platform also covers IP India registry matters. Ask for the current spec, because endpoints and limits must come from the live documentation. Public sources like IP India and eCourts offer search, but not one documented API for both.

What rate limits should I expect from a court data API?

Expect a limit per minute and a limit per month, both stated in the vendor docs. There is no standard figure in India. Ask for both in writing, ask what error you get at the limit, and design bulk jobs to stay under the per-minute cap.

How do I check the coverage of a legal API in India?

Test it with a sample of 100 to 200 matters you already know, across courts and years. Record how many are found and how many registry records return a non-null status. Always write down the sample size N and the test date, and repeat the test every month.

Does Claw track copyright registrations?

No. Claw does not yet track copyright registration with the Copyright Office. For copyright, Claw is court-side only, meaning it tracks and researches copyright litigation. For trademarks, patents, designs and GIs, it tracks IP India registry matters as well.

Can I use free public sources instead of a paid API?

For small, occasional checks, yes. IP India public search, the eCourts portal and Indian Kanoon are useful. They do not give one documented, rate-limited API across registry and courts, so for products or high-volume monitoring a licensed API is the safer route.