Revised 30 October 2026 · v1.0.0 · what changed

Code: github.com/herrrickshaw/lok-rajya-sabha-qa

Watch the walkthrough

Watch on YouTube: youtu.be/ZjVYwOQZ2to

Why it is useful

• Who benefits: policy researchers, journalists, civic-tech teams and students who want to see which subjects and ministries draw parliamentary scrutiny.
• What it replaces: searching the Lok Sabha and Rajya Sabha sites session by session and merging two different party and ministry naming systems.
• Evidence from the run: the party table sums exactly to the 59,381 questions, and one SQL query ranks the ministries.
• The repo also computes a 'lift' score for each party-and-ministry pair (how much more a party asks about a ministry than average), a quick way to see signature interests.

What it is for

The repo pulls every question since the 18th Lok Sabha convened in June 2024 from the public Parliament sources, matches each to the asking member's party and state, and aggregates by party, ministry, state and topic. It also compares scrutiny with what ministries publish through the Press Information Bureau.

This post shows aggregates only. It reports no individual MPs' records. Not political or legal advice; counts of questions say nothing about the merit of the questions.

Who it is for

• Policy and government-affairs researchers
• Journalists and civic-tech volunteers
• Students of Indian parliamentary practice

What you need

Get the code and an AI assistant

You do not have to be a programmer. Get the repository, give it to an AI assistant, and ask in plain language. Any of these routes works:

git clone https://github.com/herrrickshaw/lok-rajya-sabha-qa.git
cd lok-rajya-sabha-qa
claude      # or: gemini

Install

1. Clone the repository (needs git)

   git clone https://github.com/herrrickshaw/lok-rajya-sabha-qa.git
   cd lok-rajya-sabha-qa

2. Create a virtual environment and install DuckDB

   python3 -m venv .venv && source .venv/bin/activate && pip install duckdb

Quick start: commands and what you should see

Rank parties by questions asked

python3 - <<'E'
import duckdb
c=duckdb.connect('data/pq_ledger.duckdb',read_only=True)
for r in c.sql('select * from party_summary order by total_questions desc limit 6').fetchall(): print(r)
E

BJP 22,881, INC 11,184, DMK 3,537, AITC 2,544, SP 2,331, TDP 2,178, each with Lok Sabha and Rajya Sabha splits.

Rank ministries by questions received

python3 - <<'E'
import duckdb
c=duckdb.connect('data/pq_ledger.duckdb',read_only=True)
print(c.sql('select * from ministry_summary order by questions desc limit 6').fetchall())
E

Health and Family Welfare 3,550, Railways 2,900, Education 2,802, Agriculture and Farmers Welfare 2,754, Finance 2,253, Civil Aviation 2,129.

Sample outputs

Real output from a run on the date shown in the sources line; trimmed to fit.

Questions by party (table party_summary)

party  lok_sabha  rajya_sabha  total
BJP      11,755      11,126  22,881
INC       7,921       3,263  11,184
DMK       2,299       1,238   3,537
AITC      1,503       1,041   2,544
SP        1,959         372   2,331
TDP       1,830         348   2,178
(all parties: 34,720 + 24,661 = 59,381)

Raw counts. Parties with more members ask more questions.

Questions by ministry (table ministry_summary, 56 ministries)

HEALTH AND FAMILY WELFARE        3,550
RAILWAYS                         2,900
EDUCATION                        2,802
AGRICULTURE AND FARMERS WELFARE  2,754
FINANCE                          2,253
CIVIL AVIATION                   2,129

Ministry names are the repo's canonical forms.

Questions by state of the asking member (table state_summary)

Uttar Pradesh   6,514
Tamil Nadu      6,031
Maharashtra     5,703
Andhra Pradesh  4,072
Kerala          3,722

Counts by the asking member's state, not by the subject of the question.

Party-ministry lift (kg_party_ministry_lift, 100+ questions)

AITC  FINANCE                          166 qs  lift 1.72
BJP   COOPERATION                     459 qs  lift 1.63
SP    EDUCATION                        167 qs  lift 1.52
SP    AGRICULTURE AND FARMERS WELFARE  164 qs  lift 1.52
BJP   DEVELOPMENT OF NORTH EASTERN     102 qs  lift 1.50

Lift above 1 means the party asks about the ministry more than its overall share would suggest.

Ways to use it

Sample prompts for your AI assistant

Copy, paste and edit the parts in [brackets].

• Claude: Read README.md and query data/pq_ledger.duckdb to show the top five ministries for [party] by lift.
• Gemini: Compare Lok Sabha and Rajya Sabha question counts for the ten largest parties and explain the differences.
• ChatGPT: Here is a table of questions per ministry. Which ministries draw many questions but have few press releases?
• GitHub (Copilot Chat or Codespaces): Write a DuckDB query that shows questions per state per ministry and exports it to CSV.
• Claude: Explain the difference between starred and unstarred questions and why that matters when reading these counts.
• Any assistant: Help me write a short, neutral summary of what these tables show, with caveats about interpreting question counts.

Limits and honest caveats

Sources: Queried data/pq_ledger.duckdb of github.com/herrrickshaw/lok-rajya-sabha-qa (commit d2517ea) from a fresh clone in a throwaway virtualenv with duckdb on 19 October 2026. Aggregate tables only (party_summary, ministry_summary, state_summary, kg_party_ministry_lift). Underlying data: sansad.in and rsdoc.nic.in. Build scripts not run.

About this article: Researched, written and edited by Umashankar Triplicane Dwarakanathan, with AI research assistance; every figure is meant to trace to the primary source cited. See the Editorial Policy for how sourcing, AI use and corrections work.