More screenshots
Works with the AI Agents you already use
Buffett-Style NSE Stock Screener With DCF Value Ranges
Screen the Nifty 500 against Buffett-style quality rules on official NSE filings, get a bear/base/bull DCF value range and margin of safety, and trace every figure back to its XBRL filing. Runs on your own computer.
Free
Buffett-Style NSE Stock Screener With DCF Value Ranges
Example session with this skill installed
Is Coal India attractive on Buffett-style rules right now? Show the rules it passes and its value range.
- Read your context and instructions
- Compiled the buffett-style nse stock
Data: prices 2026-11-11 · latest FY 2026-06-30 · 399 companies scored · preset Short history (4 years)
| Company | Rules | Buffett score | Value range (bear–base–bull) | Margin of safety |
|---|---|---|---|---|
| Coal India | 8/8 | 8.1/10 | ₹234 – ₹631 – ₹1,011 | 35.6% vs ₹406.30 |
Not investment advice and not SEBI-registered advice.
Connects securely to your tools. The creator never sees your data.
About this skill
Most stock screeners show you a number. This one shows you where the number came from.
NSE Value Screener is a complete local web app that your agent installs and runs for you. It screens Indian stocks (Nifty 50 up to the Nifty 500) against documented Buffett-style rules, values each company with a bear / base / bull discounted cash flow, and shortlists the companies trading at a margin of safety. Every figure links back to the official NSE XBRL filing, tag and file hash it came from.
Real result, 12 Nov 2026
Screen: Nifty 500, Short history (4 years) preset, prices of 11 Nov 2026.
- 495 companies in the universe, 399 scored on enough validated years
- 120 passed every quality rule; 4 also traded at least 30% below the base-case estimate
- Top result: BLS International. Passes 8 of 8 rules, Buffett score 8.3/10, estimated value ₹418 (range ₹124–620) against a price of ₹229, so a 45.3% margin of safety
- 5 companies were held back as invalid data rather than scored on figures that failed validation
What you get
- Quality rules: consistent earnings, average ROE, debt/equity, margin stability, free cash flow and share dilution, with separate rules for banks and NBFCs
- Valuation as a range: a bear / base / bull scenario DCF on owner earnings attributable to the parent's shareholders, with the terminal value's share shown
- Four scores: business quality, value, data confidence, and an overall Buffett score that abstains instead of guessing when an input is missing
- Evidence for every figure: filing date, XBRL tag and context, SHA-256 of the downloaded file, and a verdict (verified, conflicting, missing…)
- Your own thresholds: edit the rules, save presets, export CSV, and keep a history of every scan
Why not just ask the AI?
Asked off the cuff, an AI quotes ratios from memory, often out of date and with no source. This app reads the official filings, checks them (units, totals, year-to-year jumps), refuses to score figures that fail, and shows an estimate as a range, never as a known value.
Private and light
Python 3.11+ standard library only: no pip packages, no API key, no account, no MCP server. Runs on 127.0.0.1. Your screens and data stay on your computer. It only downloads public NSE files, at most one request every 3 seconds, and it identifies itself honestly.
Where it stops
NSE often refuses automated clients. When it does, the app says so, and you download the files in your browser and import them offline (the skill walks you through it). A first full Nifty 500 download takes several hours. NSE's machine-readable annual results start at FY2023, so the default preset judges 4 years, not the 10+ a classic Buffett screen would use. Insurers are not scored yet.
Not investment advice and not SEBI-registered advice. It is a rules-based screen, not a prediction of Warren Buffett's own choices.
Free. If it helps your research, a short review helps other investors find it.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
Not investment advice. NSE may refuse automated clients (offline import then needed). First full download takes several hours at 1 request per 3 s. Annual XBRL history starts at FY2023 (4 years). Insurers not scored. Reconciliation re-reads the same NSE files; no BSE/annual-report second source yet.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
1 use
Times used to date
Free to use
No account required to browse
Trust & safety
Security scan
Security scan passed 4 days ago
- Free with an Agensi account
- Accounts
Nseindia
- Software
Python 3.11+
Python 3.11+ and about 2 GB of disk space. NSE often refuses automated downloads; you can then download the official files in your browser and import them offline.
Permissions
Allowed Hosts
File Scopes
Read-only HTTPS downloads of public NSE files, at most one request every 3 s. The web app binds to 127.0.0.1 only. No credentials are used or stored.