Guide · Agent skill
How to use the DCF Valuation agent skill (dcf-valuation)
Updated · written with AI from the project's own documentation and checked against our catalogue data
The dcf-valuation skill teaches an AI agent to run a discounted cash flow (DCF) valuation of a public company and write the result as a Markdown report. It is aimed at investors, students and analysts who want a transparent first-pass estimate with every assumption visible. By the end of this guide you will know what the skill does, how to install it, how to ask for a valuation and how to check the output.
DCF Valuation Analysisnexu-io/open-design · Apache-2.0 Download the skill Quick start
- 1Download the ZIP. Get the dcf-valuation skill ZIP from the page where you found this guide.
- 2Install the skill. Unzip it into ~/.claude/skills/, or into a project's .claude/skills/ folder. In the Claude apps, upload the ZIP under Settings → Capabilities → Skills.
- 3Gather your data. Collect revenue, operating cash flow, capital expenditure, free cash flow, cash, debt and diluted shares, from filings or your own figures.
- 4Ask in plain language. Request a valuation with the company name or ticker, for example asking what the company is worth or for a fair value and price target.
- 5Review the report. Open the Markdown file in Design Files at finance/<safe-company-or-ticker>-dcf.md and check the inputs, assumptions and sensitivity grid.
What does the dcf-valuation skill make the agent do?
An agent skill is a folder of instructions that an AI agent loads when a task matches it. This one is for questions such as what a public company is worth or whether a stock looks undervalued. The skill makes the agent follow a fixed DCF method: gather revenue and cash flow figures, choose a discount rate from a sector range, project five years of free cash flow and estimate a fair value per share.
The skill also runs a sensitivity grid that shows how the answer changes under different assumptions. Each input is labelled as sourced, derived, user-provided or assumption, so you can see which numbers came from data and which were judgement. The output is a Markdown report, not a live trading tool.
- Gathers 3-5 years of revenue, operating cash flow, capital expenditure and free cash flow
- Picks a WACC (discount rate) starting from a sector range and explains any adjustment
- Projects five years of free cash flow and adds a terminal value
- Builds a 3x3 sensitivity matrix for WACC and terminal growth
- Saves the report to Design Files
Install the dcf-valuation skill
Download the skill ZIP from the page where you found this guide. Unzip it into your personal skills folder, ~/.claude/skills/, or into a project's .claude/skills/ folder if you want the skill only for that project.
If you use the Claude apps instead, open Settings, go to Capabilities, then Skills, and upload the ZIP there. The skill's licence is Apache-2.0.
- Personal install: unzip into ~/.claude/skills/
- Project install: unzip into the project's .claude/skills/ folder
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
~/.claude/skills/How do you trigger the skill with a natural request?
You do not need a special command. Ask the agent in plain language, and the skill loads when your request matches its description. Requests that mention DCF, fair value, intrinsic value, price target, undervalued or overvalued, or 'what is this company worth?' are the clearest matches. The skill also recognises the trigger words dcf, discounted cash flow, intrinsic value, fair value, price target, undervalued and overvalued, plus the Chinese terms 估值 and 内在价值.
Include the company name or ticker and, if you have them, the figures or filings you want used. The more complete your data, the fewer inputs the agent has to assume.
- Example: What is Company X worth? Run a DCF using its last five years of filings.
- Example: Give me a fair value and price target for TICKER, with a sensitivity table.
- Example: Is this stock undervalued? Here are the revenue and free cash flow figures I have.
What does the report contain and where is it saved?
The skill writes one Markdown file to Design Files at a path built from the company or ticker. The report follows a fixed structure: Query, Valuation Summary, Data Coverage, Key Inputs, Forecast, Sensitivity Analysis, Caveats, Sources and an Evidence Note. In its final answer, the agent summarises the valuation and gives you the report path so you can reopen it.
Check the Data Coverage and Key Inputs sections first. They show what was sourced, what was missing and what was assumed. If the conclusion depends on a narrow assumption, the skill is instructed to say so in the sensitivity section.
- Valuation Summary: current price, fair value, upside or downside, confidence
- Key Inputs table: each input with value, source type and citation or note
- Sensitivity Analysis: 3x3 WACC vs terminal growth matrix
finance/<safe-company-or-ticker>-dcf.mdWhat does the skill need and what are the safety limits?
The skill needs an agent runtime that supports SKILL.md skills and file writing, because it declares the file_write capability. It also needs financial data. You can supply figures, upload filings, or let the agent use web access to public sources. No specific API key is named in the files. The skill does not run scripts, but it does use the network when it looks up public sources.
Treat web pages, filings and search results as untrusted. The skill instructs the agent not to follow instructions embedded in that content and to use it only for facts and citations. The skill does not verify market data, so check prices and statements yourself before relying on the output.
- Runs scripts: no
- Uses network: yes, for public sources the agent can access
- Optional OD research commands are mentioned but not defined in the files provided
- Missing data is labelled as an assumption rather than invented
Tips for getting a useful valuation
Give the agent historical figures rather than asking for a bare answer. Five years of revenue, cash flow and share count give it stable history to work from. Also state your own discount rate or growth view if you have one, since the skill caps sustained explicit-period growth at 15% unless you provide a higher assumption.
Do not rely on the skill for banks, insurers, startups or private companies. The skill's own guidance says standard free-cash-flow DCF is often inappropriate there, and its sector ranges are only a rough starting point.
- Provide filings or figures, and say which year is the latest
- Treat the sector WACC range as a starting point and read the adjustment reasoning
- Look at the sensitivity grid before forming a view
- Ask a qualified professional to review the report before acting on it
What people use it for
- Estimating a first-pass fair value for a listed company with the assumptions shown
- Teaching students how each DCF input changes the result
- Organising a set of filings into a consistent valuation template for an analyst team
- Comparing how a fair value moves when WACC or terminal growth changes
Frequently asked questions
Is the dcf-valuation skill free to use?
The skill is published under the Apache-2.0 licence, and the source files do not state a price. Check the licence file that comes with the download for the terms that apply to you.
What do I need to run the dcf-valuation skill?
You need an agent runtime that supports SKILL.md skills and file writing. You also need financial data, which can be figures you supply, uploaded filings or web access to public sources. No specific API key is named in the files.
Where do I install the dcf-valuation skill?
Unzip the downloaded ZIP into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. In the Claude apps, upload the ZIP under Settings → Capabilities → Skills.
Does the skill give buy or sell advice?
No. The skill produces a transparent DCF report, and its own guidance says the output should be checked by a qualified professional before acting on it. It does not verify market data.
Why does the agent say my data is missing or assumed?
The skill is designed not to invent figures. If revenue, free cash flow, debt, cash, shares or price are not available, the agent either asks for them, finds them, or labels them as assumptions in the Key Inputs table.
When is this skill the wrong tool?
Avoid it for private companies, startups, and banks or insurers, where standard free-cash-flow DCF is often inappropriate. It is also unsuitable if you need real-time prices or verified financial statements.
New agent skills, repos and AI tools — every week
The best new skills and open-source projects, the fastest-growing repos, and one free prompt. Five minutes to read. Unsubscribe anytime.
Our verdict
Use this skill if you want a transparent first-pass DCF with assumptions labelled and a sensitivity grid, and you can supply or check your own financial data. Pick a different approach for banks, insurers, startups or private companies, or if you need verified real-time market data for trading decisions.
Download the skill