Guide · Agent skill
How to use the Scenario War Room agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The Scenario War Room is an agent skill for what-if planning when more than one bad thing might happen at once, such as losing a key customer while a fundraise slips. It is written for founders, executives and leadership teams who want a structured planning exercise instead of ad hoc brainstorming. After this guide you will know what the skill does, how to install it, how to phrase a request that triggers it, and what to check before acting on its output.
Scenario War Roomalirezarezvani/claude-skills · MIT Download the skill Quick start
- 1Download the ZIP. Download the scenario-war-room ZIP from this page.
- 2Unzip into a skills folder. Unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder.
- 3Or upload in the Claude apps. Go to Settings, then Capabilities, then Skills, and upload the ZIP.
- 4Describe a compound risk. Ask a what-if question naming two or three variables, such as losing a top customer while a fundraise slips.
- 5Review the severity levels. Check the base, stress and severe outputs and the cascade map to see how one problem triggers the next.
- 6Set up early warning signals. Monitor the listed trigger points and act on the hedges with their owners and deadlines.
What is an agent skill, and what does this one make the agent do?
An agent skill is a folder of instructions (a SKILL.md file, plus optional reference files and scripts) that an AI agent loads when a task matches the skill's description. This skill tells the agent to run a structured cascade model rather than a quick answer.
Unlike a single-assumption stress test, the skill models compound adversity across all business functions at once. It assigns each variable a probability and timeline, maps how one problem triggers consequences in finance, hiring, engineering and other functions, and rates base, stress and severe cases. It ends with early warning signs and hedging actions that have an owner and a deadline.
- Limits scenarios to a maximum of 3 variables
- Maps cascade effects across roles such as CFO, CRO, CTO, CHRO and COO
- Rates Base (one variable), Stress (two) and Severe (all) cases
- Lists measurable trigger points that signal a scenario is unfolding
- Produces hedges with cost, impact, owner and deadline
Install the Scenario War Room skill
Download the scenario-war-room ZIP from this page and unzip it. For a personal install, place the unzipped folder in ~/.claude/skills/. For a single project, place it in that project's .claude/skills/ folder.
You can also upload the skill in the Claude apps. Go to Settings, then Capabilities, then Skills, and upload the ZIP there.
- Personal install: ~/.claude/skills/
- Project install: .claude/skills/ inside your project
- Claude apps: Settings → Capabilities → Skills, then upload the ZIP
Trigger the skill with natural requests
You do not need a special command. The skill is meant to load when you describe a compound risk, a strategic decision with major downside, or ask a 'what if X AND Y both happen?' question. Name the variables and, where you can, attach rough numbers and timelines.
The source gives these examples of requests that fit the skill.
- What if we lose our top customer AND miss the Q3 fundraise?
- What if 3 engineers quit AND we need to ship by Q3?
- What if our market shrinks 30% AND a competitor raises $50M?
Use the optional Python script for an interactive build
The skill bundles scripts/scenario_modeler.py, an interactive scenario builder with cascade modeling. It is optional. The agent's output is planning text, and the script is a separate way to structure the same exercise. The script uses only the Python 3 standard library.
Run it from the skill folder with the command below.
python scripts/scenario_modeler.pyRead the output and check the assumptions
Each session produces a fixed set of sections: scenario name, variables, most likely path, severity levels, cascade map, early warning signals, hedges and a recommended decision. Treat the probabilities as guesses unless you supply real data. The example figures such as ARR, runway and costs are illustrative.
The skill separates the base case from sensitivity and models recovery as well as impact. Use its output to structure a discussion, not as a final financial forecast.
- Replace illustrative numbers with your own ARR, runway and cost data before relying on them
- Ask a finance specialist to review any precise financial model
- Get professional review for legal, compliance or investment decisions
Safety notes: scripts and network
The skill runs scripts: yes, because scripts/scenario_modeler.py is bundled. It uses network: no, according to the skill's listing. Read the script before running it, as you would any script you download.
- Scripts: yes (scripts/scenario_modeler.py)
- Network: no
- License: MIT
Tips for better war room sessions
Keep to three variables per scenario. The skill itself warns that more than three is noise. Quantify where you can, since 'revenue drops' is less useful than a specific ARR figure at risk over a set period. Run three or four scenarios per planning cycle rather than dozens.
- Pick the three variables that actually worry you
- Give ranges when you are uncertain
- Ask for the cascade, not just the first-order hit
- Ask for recovery actions, not only impact
- Use it alongside a pre-mortem or contingency session with your leadership team
What people use it for
- A founder modelling a key customer churning while a Series A close slips
- A leadership team running a pre-mortem on a major strategic bet
- A CEO preparing contingency plans for a key engineer leaving during a product deadline
- An executive team defining early warning signals for a revenue miss combined with a competitor move
Frequently asked questions
How do I install the scenario-war-room skill?
Download the ZIP from this page and unzip it into ~/.claude/skills/ for personal use, or into a project's .claude/skills/ folder. You can also upload the ZIP in the Claude apps under Settings, then Capabilities, then Skills.
What do I need to run the skill?
The skill requires Python 3 only for the optional script, which uses only the standard library. The agent's main output is planning text, so the script is not required to use the skill.
Is the scenario-war-room skill free?
The skill is licensed under MIT, as stated in its listing. Check the license file in the download for the full terms.
Does the skill access the internet?
The listing says it uses network: no. It does run scripts, because it bundles scripts/scenario_modeler.py, so read that script before running it.
How is this different from a stress test?
A single-assumption stress test changes one input. This skill models compound adversity across functions at once, and the skill itself says single-assumption stress testing is a different tool.
Why are the probabilities and numbers not reliable?
The agent's probabilities are guesses unless you supply data, and the example figures for ARR, runway and costs are illustrative. Replace them with your real company numbers before relying on the output.
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 are a founder or leadership team who wants a structured template for compound what-if risks and hedging actions. Pick something else if you need single-assumption sensitivity analysis, a precise financial model built on real data, or a decision on legal, compliance or investment matters without professional review.
Download the skill