More screenshots
Works with the AI Agents you already use
Valuation Triangulator: DCF, reverse DCF and multiples
Value a stock three ways, weight bear/base/bull, and treat model disagreement as the signal. Offline script included.
$7
Valuation Triangulator: DCF, reverse DCF and multiples
Example session with this skill installed
Value this stock three ways. FCF 100, 5-year growth 8%, discount rate 9%, 50 shares, net debt 200, price 40, EPS 3, own-history P/Es 18/20/22/25/21, peer P/E 19. (Illustrative figures.)
- Read your context and instructions
- Compiled the valuation triangulator
$ triangulate.py --fcf 100 --growth 0.08 --discount 0.09 --shares 50
--net-debt 200 --price 40 --eps 3 --pe-hist 18,20,22,25,21
--pe-peer 19 --hist-growth 0.06 (illustrative inputs, run 2026-11-16)
- DCF (growth cases, same discount and terminal)
bear growth 3.0% value 28.23 terminal share 74% weight 25%
base growth 8.0% value 35.85 terminal share 76% weight 50%
bull growth 13.0% value 44.92 terminal share 77% weight 25%
scenario-weighted value 36.21 (vs price -9.5%) - Reverse DCF
price 40.00 implies 10.4% FCF growth for 5y (+4.4 pts vs the record) - Multiples
eps used 3.00 current P/E 13.3
own history: median P/E 21.0, today at the 0th percentile -> value 63.00
peers: value 57.00
Margin of safety (value - price) / value
dcf_weighted -10.5%
multiple_own_history +36.5%
multiple_peers +29.8%
DISPERSION DIVERGENT (max-min)/median = 47% methods disagree on direction
to match multiple_own_history, the DCF needs 21.0% growth
to match multiple_peers, the DCF needs 18.6% growth
Connects securely to your tools. The creator never sees your data.
About this skill
A DCF says 36, the P/E history says 63, the price says 40. Which one do you believe? This skill's answer: none of them yet. It values a stock three ways, keeps the methods apart, and treats the gap between them as the finding instead of averaging it into a comfortable number.
What it does
Your agent runs a DCF with bear/base/bull growth cases, a reverse DCF that solves (by bisection) for the growth today's price already assumes, and multiples against the company's own P/E history and its peers, with EPS restated first if margins sit at a cycle peak. It reports terminal-value share, margin of safety as (value - price) / value, and a dispersion flag. When methods diverge it prints the growth the DCF would need to match each multiple, so you can see which assumption is unusual. Rules cap explicit growth, bound terminal growth, and refuse inputs that have no answer.
What you get
| Path | Holds |
|---|---|
| SKILL.md | eight rules, which method leads by company type, five-step method |
| references/dcf-inputs.md | firm vs equity FCF, SBC, discount rate, share and currency traps |
| references/reverse-dcf.md | reading implied growth against the record |
| references/multiples-history.md | rebuilding P/E history, peak normalisation, peers, PEG |
| references/scenarios.md | case design, weights, expected return |
| templates/valuation-memo.md | copy-paste memo with placeholders |
| scripts/triangulate.py | offline DCF, reverse DCF, multiples, dispersion (stdlib) |
| tests/test_triangulate.py | 12 offline tests, happy paths and refusals |
Who it's for
Self-directed investors, analysts and developers who ask an agent what a stock is worth and want the assumptions on the table, not one confident number.
Scope
Covered: free-cash-flow companies valued by DCF, reverse DCF and P/E multiples; scenario weighting; guidance on which method leads for cyclicals, banks, REITs, utilities and loss-makers.
Not covered: fetching prices or filings (you supply the inputs), a bank or REIT model in the script, price targets. Script thresholds (75% terminal share, 25% and 50% dispersion bands) are the skill's defaults, not market law.
This is analysis tooling, not investment advice.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse All Claude Skills.
The script runs a single-stage five-year DCF with a Gordon terminal value and P/E multiples only. It does not model banks, REITs or loss-makers, fetch data, or choose growth and discount rate for you. Thresholds are defaults, not market law. Valuation illustrates assumptions; it is not a price target.
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
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 1 day ago
- Software
Python 3.9+Your own company figures
You supply the inputs: free cash flow, share count, net debt, price and P/E history. The script fetches nothing.
Permissions
The bundled script is offline: no network, no credentials, no files written. It reads a JSON input file only if you pass --input.