Check which Indian benefit programs fit a person
An agent compares a person’s details with scheme rules and returns likely matches from Indian government programs.
- What they did
- A developer built a custom workflow on Inception’s Mercury with the Vercel AI SDK and connected it to Sanity through a Context MCP setup using GROQ. It extracts the user’s profile, reads the schema, and makes a single query for programs that have typed eligibility rules, then evaluates each exclusion rule. After that, it formats the matches into JSON cards and appends the apply and source links stored in the dataset.
- How it went
- It supports English and Kannada and can handle follow-up questions. The dataset holds 125 schemes, but only 10 flagship programs include structured rules, with 33 rules and 11 exclusions; each run takes roughly 30 to 45 seconds. The author did not audit every rule manually, and an earlier Knowledge Base route used 20 tool calls without answering before GROQ mode corrected the issue.
- Tips
- Give the agent concrete details such as state, income, household makeup, and any category, land, or disability information. Because only a small slice of the dataset has structured rules and a deployed Studio is required, verify the official application pages before acting on the output.
Try it yourself with Perplexity Comet
I live in [state] in India. I am [age], [occupation], and my household income is about [amount]. Relevant details include [caste or category], [land ownership or use], [family size], [disability status], [student status], and [other facts]. Compare my information with official government scheme data and list the central and state programs I could qualify for. For each possible match, include the scheme name, the benefit, the key eligibility points, the documents usually requested, and the official application or source link. If a rule is unclear or incomplete, say so. Before any action that would submit, send, delete, pay, or share my information, stop and ask for my approval. Do not submit applications, enter ID numbers, or carry out irreversible steps.
Fill in the [brackets]. Agents can make mistakes, so review anything before it is sent, paid or deleted.
Get 3 agent uses like this every morning
With the day's top AI news. Free, and one click to unsubscribe.
Related agent uses
Find billing mismatches in contractor invoices
They used an AI to compare a contractor’s paperwork and draft dispute emails that uncovered pricing and change-order problems.
Have an agent file insurance overcharge claims
Tarun used an agent to submit claims after finding insurance overcharges on services that should have cost nothing out of pocket.
Estimate care costs and negotiate medical bills
An app routes people to the right care setting, predicts insurance-based visit costs, and can dispute US medical bills afterward.
Use a chat agent to push for a lower support quote
An AI agent handled a GoDaddy support chat and secured a cheaper price after the person set the rules and waited for the result.
Use an agent to push a warranty claim by email
A user let an agent handle a mattress warranty thread, pressing the company with legal points until the back-and-forth stalled.
Prepare overseas supplier payments for review
An approved agent checked account balances and assembled supplier payment details so a person could review them before paying.