How to use the Pricing Strategist agent skill (Claude)

What the Pricing Strategist skill makes the agent do

The skill is built around three questions asked at the pricing-design moment: which pricing model fits your product, customer and market; what customers will pay before a price feels too expensive; and how to package the product into Good, Better and Best tiers.

It recommends a model and a price range with trade-offs, never a single final number. The source is explicit that a person still picks the number, owns the trade-offs and runs the go-to-market plan.

• Pricing model selection: ranks five models (subscription seat-based, usage-based, value-based, freemium, hybrid) with a fit score from 0 to 100 and trade-offs.
• Van Westendorp analysis: reads survey answers (too cheap, bargain, getting expensive, too expensive) and returns four intersection points and a Range of Acceptable Prices.
• Packaging design: assigns features to three tiers and flags common anti-patterns such as a decoy tier, a feature dump in Best, or no upgrade trigger.

Install the pricing-strategist skill

Option 1 is a manual install. Download the ZIP of the skill from the page you are reading, then unzip it into your personal skills folder ~/.claude/skills/ or into a project's .claude/skills/ folder so the skill is available only inside that project.

Option 2 is to upload the ZIP in the Claude apps. Open Settings, go to Capabilities, then Skills, and upload the file there.

The skill is MIT licensed, and its folder contains SKILL.md, three scripts in scripts/, three reference documents in references/ and a brief template in assets/.

• Personal install: ~/.claude/skills/ (available across your projects).
• Project install: .claude/skills/ inside the project folder.
• Claude apps: Settings → Capabilities → Skills, then upload the ZIP.

Check what you need before you start

The scripts need Python 3. They use only the standard library, so there are no third-party packages to install. The skill also accepts optional inputs: a filled-in pricing brief and Van Westendorp survey data in JSON.

You can confirm each script runs by calling it with --help, and see the built-in example by calling it with --sample. The source says both flags work on all three scripts.

• Python 3 installed on the machine where the agent runs the scripts.
• Optional: a completed pricing brief based on assets/pricing_brief_template.md.
• Optional: Van Westendorp survey responses saved as JSON, with at least four questions per respondent.

Trigger the skill with natural requests

You do not need a special command. Describe the pricing moment in plain words and the agent should match the request to the skill. The skill's description covers designing or revisiting pricing, selecting a model, running Van Westendorp analysis and designing Good/Better/Best tiers.

Give the agent the context it needs: your product type, customer, average deal size, and whether you have survey data. The source says it works best when the inputs are specific and unknown fields are marked as unknown rather than guessed.

• 'Help me choose a pricing model for our new API product. Here is our brief.'
• 'Run a Van Westendorp analysis on this survey file and tell me the acceptable price range.'
• 'Draft Good, Better and Best tiers for these features and flag any packaging anti-patterns.'
• 'We have 18 months of sales data and are considering moving from seats to usage. Walk me through the trade-offs.'

Run the scripts yourself

The skill's workflow runs three scripts in order. The model picker comes first because tier structure depends on the chosen model. The WTP analyser and packaging designer can then be run with your own files.

The built-in example runs from the skill's folder in the repository. Once installed, your folder path will differ.

• scripts/pricing_model_picker.py --input brief.json --profile saas --output markdown (ranks the five models).
• scripts/wtp_analyzer.py --input survey.json --output markdown (Van Westendorp intersection points and range).
• scripts/packaging_designer.py --input features.json --profile saas --output markdown (Good/Better/Best assignment and flags).

Safety notes: scripts and network access

The skill runs scripts, so read them before you run them on sensitive data. The source states that the skill runs scripts and does not use the network. Your pricing brief and survey data stay in the files you pass in, as far as the source describes.

Sample size matters for the survey method. The skill treats Van Westendorp results as directional, requires at least 30 respondents as a minimum and prefers 100 or more. Below 30 the script emits a sample-size warning, and the source says to respect it.

• Runs scripts: yes. Review scripts/ before running them.
• Uses network: no, according to the skill's metadata.
• Do not use the skill for approving individual customer discounts. That is a separate deal-desk function.

Tips for getting useful output

Fill in the pricing brief before running the model picker. The source says the outputs are only as good as the inputs, so mark gaps as unknown instead of guessing.

Treat Van Westendorp output as a range to test in market, not a price. Value-based pricing needs a measurable customer value metric. If you cannot measure customer value, the source advises against choosing value-based pricing.

• Use the brief template to capture industry, average deal size, customer count, value drivers and competitor models.
• Keep Best tier feature growth modest relative to price. The anti-pattern guidance warns against Best having much more features for much higher price.
• Avoid mixing usage charges into a subscription tier. The source calls this two pricing models in one.