Property Legal Opinion API from your documents

Example

Request

curl -X POST https://verified.realestate/api/v1/tools/legal_opinion_project \
  -H "Authorization: Bearer $LANDLENS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"action":"create","survey_numbers":["65/4B"],"lat":12.8421,"lng":80.031,"district_code":"03","taluk_code":"14","village_code":"196","village_name":"Kavanoor","rural_urban":"rural","upload_ids":["up_lcja6mbi1t"]}'

Responsereal answer

{
  "ok": true,
  "status": "proposal",
  "data": {
    "proposal": true,
    "action": "legal_opinion_project",
    "params": { "name": "Legal opinion - Survey 65, Kavanoor", "survey_numbers": [{ "number": "65", "subdivision": "4B", "isPrimary": true }], "...": "..." },
    "quote": {
      "credits": 10,
      "lines": ["10 credits for the opinion (up to 100 pages)"],
      "pages": 1,
      "documents": 1,
      "survey_numbers": 1
    },
    "documents": [
      { "upload_id": "up_lcja6mbi1t", "file_name": "sample-sale-deed.pdf", "pages": 1, "size_kb": 1, "uploaded_at": "2026-10-22T21:38:36+05:30" }
    ],
    "registration_office": { "sro": "Tambaram Joint 1", "village": "KAVANOOR" },
    "what_happens": "On approval 10 credits are billed up front and the documents are read, classified and cross-checked against the land registers; the AI opinion then generates. Expect one to several hours. ...",
    "message": "Legal opinion on 1 document(s), 1 page(s), survey 65/4B: 10 credits."
  },
  "credits_quoted": 0,
  "credits_charged": 0,
  "next": { "tool": "approve_action", "arguments": { "action": "legal_opinion_project", "params": { "...": "..." } } }
}

A real quote from the API on a one-page test file, trimmed for length. It was not approved.

What it returns

Six free steps around one paid approval. upload and create are free; create returns the quote with status proposal; approve_action starts the opinion; status and answer follow it.

Field	What it holds
`upload (upload)`	upload_id, file_name, pages and size_kb for the file kept, how many uploads are waiting and their total pages. Files wait 24 hours.
`quote (create)`	credits, the lines that make it up, pages, documents and survey_numbers.
`documents, registration_office, what_happens (create)`	The files the opinion will read, the sub registrar office worked out from the revenue codes, and what happens on approval.
`project_id, credits_used, project_url (approve_action)`	The project created, what it cost, and its page in LandLens.
`status (status)`	The processing stage and progress, whether a report is ready, what is pending (missing documents, questions) and next, the step to take now.
`recorded, submitted, pending (answer)`	What your answer recorded, whether the opinion resumed, and anything still open.

Ways to call it

The steps, in order. Every step but the approval is free.

1. Upload each document

One file per call, base64. A file over about 900 KB goes in numbered parts: part 1 with file_name and parts, then the rest with the upload_id part 1 returned.

{"action": "upload","file_name": "sale-deed-2011.pdf","file_base64": "JVBERi0xLjQK..."}

2. Create, and get the quote

The location, the survey numbers and the revenue codes (resolve_parcel gives them). Leave out upload_ids to use every waiting file.

{"action": "create","survey_numbers": ["65/4B"],"lat": 12.8421,"lng": 80.031,"district_code": "03","taluk_code": "14","village_code": "196"}

3. Approve

Send the next block to approve_action unchanged. This is where the opinion is charged.

{"action": "legal_opinion_project","params": {"...": "..."}}

4. Follow it

Poll status. If it asks for a document, upload it and attach it with answer; answer its questions the same way, then submit_all.

{"action": "status","project_id": 101}

Inputs

The JSON body of the POST is the arguments. Only action is required by the schema; each step takes the inputs shown in its row above.

Argument	Type	What to send
`action`required	string: upload, uploads, remove, create, status, answer	"upload", "uploads" (list waiting files), "remove", "create" (the quote), "status" or "answer".
`file_name`	string	upload: the file name with its extension, e.g. sale-deed-2019.pdf
`file_base64`	string	upload: the file, base64 encoded, at most about 900 KB per call. Bigger files go in parts.
`part`	integer	upload: this part number, 1-based, when a file is sent in parts
`parts`	integer	upload: how many parts the file is sent in (omit for a single-call upload)
`upload_id`	string	remove: the upload_id to drop; upload parts 2..N: the upload_id part 1 returned
`survey_numbers`	array	create: the survey numbers, e.g. ["65/4B"] or [{"number": "65", "subdivision": "4B"}].
`lat`	number
`lng`	number
`district_code`	string	create: revenue codes from the free parcel lookup; they let the register checks run without guessing
`taluk_code`	string
`village_code`	string
`rural_urban`	string: rural, urban
`upload_ids`	array	create: which waiting files to use. Leave out to use them all.
`project_id`	integer	status and answer: the project from the approval, or from legal_opinion list.
`document_actions`	object	answer: keyed by the missing document id from status; each {action: uploaded\|not_available\|explained\|identifier_provided, upload_id?, value?}
`question_answers`	object	answer: keyed by the question id from status; the answer text
`submit_all`	boolean	answer: true to hand the answers to the pipeline and continue
`attach_upload_ids`	array	answer: pending uploads to add to the project as further documents when they answer no particular request
`district_name`	string
`firka_ward_number`	string
`is_merged_upload`	boolean	upload: true when one file holds several documents scanned together
`name`	string	create: a name for the project (optional)
`questionnaire`	object	create: what the workspace asks: dealer_name, dealer_relationship, property_type, cersai_property_type, area_value, area_unit, is_vacant, concerns
`sro_code`	string	create: the sub-registrar office code from the free SRO lookup or the EC in hand; used when the revenue codes do not map to an office, common in Chennai
`sro_name`	string
`survey_numbers_uncertain`	boolean
`taluk_details_uncertain`	boolean
`taluk_name`	string
`urban_block_number`	string
`village_name`	string

Code

Set LANDLENS_KEY to your key. In an AI app, connect https://verified.realestate/mcp once and ask in plain words instead. In Postman, download this API's collection, import it (File, Import) and set apiKey.

curl -X POST https://verified.realestate/api/v1/tools/legal_opinion_project \
  -H "Authorization: Bearer $LANDLENS_KEY" \
  -H "Content-Type: application/json" \
  -d '{"action":"create","survey_numbers":["65/4B"],"lat":12.8421,"lng":80.031,"district_code":"03","taluk_code":"14","village_code":"196","village_name":"Kavanoor","rural_urban":"rural","upload_ids":["up_lcja6mbi1t"]}'

import os, requests

r = requests.post(
    "https://verified.realestate/api/v1/tools/legal_opinion_project",
    headers={"Authorization": f"Bearer {os.environ['LANDLENS_KEY']}"},
    json={
        "action": "create",
        "survey_numbers": [
            "65/4B"
        ],
        "lat": 12.8421,
        "lng": 80.031,
        "district_code": "03",
        "taluk_code": "14",
        "village_code": "196",
        "village_name": "Kavanoor",
        "rural_urban": "rural",
        "upload_ids": [
            "up_lcja6mbi1t"
        ]
    },
    timeout=60,
)
body = r.json()
print(body["status"], body["credits_charged"], body["data"])

const res = await fetch("https://verified.realestate/api/v1/tools/legal_opinion_project", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.LANDLENS_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "action": "create",
    "survey_numbers": [
      "65/4B"
    ],
    "lat": 12.8421,
    "lng": 80.031,
    "district_code": "03",
    "taluk_code": "14",
    "village_code": "196",
    "village_name": "Kavanoor",
    "rural_urban": "rural",
    "upload_ids": [
      "up_lcja6mbi1t"
    ]
  }),
});
const body = await res.json();
console.log(body.status, body.credits_charged, body.data);

Here are the sale deed, parent deed and patta for survey 65/4B in Kavanoor. Get me a legal opinion on them.

Your AI uploads the files, shows you the itemised quote, and starts the opinion only after you say yes.

Responses

Every answer has the same envelope. What status can say for this call:

status	What it means
`ok`	upload, uploads, remove, status or answer completed; or, from approve_action, the project was created and is processing.
`proposal`	From create: the quote. Nothing is charged. Approve it by sending next.arguments to approve_action within 30 minutes, unchanged.
`not_found`	status for a project id that is not on your account.
`invalid_args`	A step was missing what it needs (no file, no survey numbers, a part out of order, a file over 25 MB), or the waiting files behind a quote have expired. error says which.
`insufficient_credits`	HTTP 402 at approval, with the credits needed, your balance and where to top up.

Where the data comes from

Register

The documents you upload, read page by page, cross-checked against the revenue records, the registration department's encumbrance index and the other registers the LandLens pipeline reads, and written into one opinion.

Coverage

Tamil Nadu property, village or town survey. PDF, JPEG, PNG, DOC and DOCX files up to 25 MB each, up to 25 files waiting for one project.

How current

The registers are read when the opinion runs, so the cross-checks are as of that day.

How we read it

The same pipeline as the LandLens workspace: documents are classified, the title chain reconstructed from them and checked against the registers, and anything missing becomes a question or a document request you answer before the opinion is finished.

Who uses it

Banks and NBFCs

Send the borrower's title documents from your loan system and get a verdict, risk score and conditions back on the file.

Law firms

Get a first reading of a bundle of deeds, with the reconstructed chain and the gaps, before the advocate's own review.

Property platforms

Offer a title opinion on the documents a seller uploads, inside your own flow.

Limits and good practice

One file per call, about 900 KB of file per call. Larger files go in numbered parts; each document can be up to 25 MB.
Waiting files are kept 24 hours. A quote whose files have expired is refused at approval; upload again and create again.
The opinion reads what the documents and registers show. Points it lists as needing manual verification have not been verified, and it is not a substitute for an advocate's signed opinion where one is required.
The opinion can pause and ask for a missing document or an answer. Poll status and reply with answer, or it waits.

Pricing

10 credits on approval

An opinion costs from 10 credits for up to 100 pages, with more for each further 50 pages and for each survey number beyond three, all shown in the quote. It is charged once, at approval; every other step is free. Credits come from the same LandLens wallet as the chat and the minis. How billing works.

Calling at volume?

For thousands of calls a month, a portfolio sweep or a platform built on these registers, we quote a customised bulk price, with invoicing to match.

WhatsApp 99961 29996

Questions

Is there an API for a property legal opinion in India?

Yes, for Tamil Nadu property. Upload the documents to LandLens over the API, approve the quote, and read the finished title opinion as JSON: verdict, risk score, title chain, red flags and recommendations.

How much does it cost?

From 10 credits for up to 100 pages and three survey numbers. Longer files and more survey numbers add to it, and the quote shows the exact total before you approve.

How long does it take?

One to several hours after approval. Poll status for the stage and progress.

How do I send a large PDF?

In parts: part 1 with file_name and parts, then each further part with the upload_id part 1 returned, each at most about 900 KB of file.

How do I read the finished opinion?

With legal_opinion, action read and the project_id.