Guide · Agent skill
How to use the CPO Product Strategy Advisor agent skill
Updated · written with AI from the project's own documentation and checked against our catalogue data
The CPO Advisor is an agent skill that gives an AI agent a chief product officer's playbook for strategic product decisions: product vision, which products to invest in, maintain or kill, product-market fit, and product team structure. It is aimed at founders, product leaders and product managers at scaling companies. After this guide you will know what an agent skill is, how to install this one, how to trigger it with plain requests, and how to run its two Python scripts on your own data.
CPO Product Strategy Advisoralirezarezvani/claude-skills · MIT Download the skill Quick start
- 1Download the ZIP. Get the cpo-advisor ZIP from the directory page.
- 2Unzip into a skills folder. Extract it to ~/.claude/skills/ for personal use, or to a project's .claude/skills/ folder.
- 3Or upload in the Claude apps. Go to Settings → Capabilities → Skills and upload the ZIP.
- 4Check the folder contents. Confirm SKILL.md, references/ and scripts/ are present after unzipping.
- 5Ask a strategy question. Describe a product vision, portfolio, PMF or org decision in plain language.
- 6Run the scripts if needed. Install Python 3 and run python scripts/pmf_scorer.py or python scripts/portfolio_analyzer.py from the skill folder.
What is an agent skill and what does cpo-advisor do?
An agent skill is a folder of instructions and reference files that an AI agent loads when a request matches the skill's description. The skill's main file, SKILL.md, tells the agent when to use it and how to behave. The skill's reference files hold the detailed material the agent reads when needed.
When you ask about product vision, portfolio strategy, product-market fit, product organisation or roadmap strategy, the cpo-advisor skill makes the agent work from a chief product officer's framework. It draws on three reference guides: a PMF playbook, a product strategy reference and a product org design reference. It is strategic guidance, not feature-level work. The skill explicitly excludes day-to-day ticket work.
- Product vision and where the product is going, using Jobs to Be Done and positioning frameworks
- Portfolio decisions: invest, maintain or kill for each product
- Measuring and protecting product-market fit, including a retention-based scorer
- Product org design: team structure, PM ratios and the product trio
- Board-level product reporting and north star metric guidance
How do you install the cpo-advisor skill?
You can install the skill in one of two ways. For Claude Code, download the ZIP from the directory page and unzip it into your personal skills folder at ~/.claude/skills/, or into a project's .claude/skills/ folder if you want it only for that project. In the Claude apps, you can instead upload the ZIP under Settings → Capabilities → Skills.
After unzipping, the folder should contain SKILL.md, the references/ folder and the scripts/ folder. The skill is then available to the agent without further configuration.
- 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
How do you trigger the skill with natural requests?
You do not need a special command. The skill's description lists the situations it covers, so the agent will usually pick it up when your request matches. Mentioning terms such as CPO, product strategy, product-market fit, product organisation, portfolio prioritisation or roadmap strategy helps.
Ask in plain language about the decision you are facing rather than naming the skill. The skill's own examples of outputs show what you can expect back.
- "Do we have product-market fit? Here is our retention data."
- "Evaluate our product portfolio and tell me what to invest in, maintain or kill."
- "Prioritise our roadmap at the portfolio level."
- "Design our product org for the next 12 months."
- "Help me write the product section of our board report."
How do you run the PMF scorer and portfolio analyzer?
The skill includes two Python scripts. pmf_scorer.py scores product-market fit across retention, engagement, satisfaction and growth. portfolio_analyzer.py classifies a product portfolio using a BCG-style matrix and gives invest, maintain or kill recommendations along with a portfolio health score. Run them from the skill folder with Python 3. Without your own data, they use built-in sample data.
To use your own numbers, supply product data in JSON format. The source confirms this option but does not show the JSON schema, so check the script's input handling in the scripts folder before preparing your file.
- Python 3 is required
- Scripts use only standard-library imports (json, sys, argparse, math, typing)
- Built-in sample data runs without any input file
python scripts/pmf_scorer.py
python scripts/portfolio_analyzer.pyWhat are the safety notes for the scripts?
The skill runs scripts but does not use the network. The scripts use simplified weights and sample thresholds, so their outputs are a starting point for discussion, not a verdict.
Do not make irreversible kill or layoff decisions from script output alone. Use the scores alongside real customer research and your own knowledge of the business.
- Runs scripts: yes. Uses network: no.
- Scores are generic and do not replace real customer research.
- Treat kill and layoff recommendations as prompts for further review.
What tips help you get useful results?
Give the agent real figures and context. The skill's diagnostic questions, such as D30 retention by cohort and the share of users who would be very disappointed if the product disappeared, show what kind of data the framework relies on. Vague requests get generic answers.
Use the north star hierarchy the skill describes: one north star metric owned by the CPO, leading indicators owned by PMs, and lagging indicators such as revenue, churn and NPS.
- Bring retention curves and cohort data for PMF questions
- List every product in the portfolio when asking for invest, maintain or kill calls
- State your business model so the north star suggestion fits
What people use it for
- Deciding which products in a portfolio to invest in, maintain or kill
- Checking whether retention and engagement data support a claim of product-market fit
- Preparing a product section for a board report
- Designing product team structure and PM ratios for a growing company
- Choosing a north star metric for a B2B SaaS, consumer or marketplace product
Frequently asked questions
How do I install the cpo-advisor skill?
Download the ZIP from the directory page and unzip it 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.
What do I need to run the cpo-advisor scripts?
You need Python 3. The scripts use only standard-library imports such as json, sys, argparse, math and typing. They include built-in sample data, and you can supply your own product data in JSON format.
Is the cpo-advisor skill free?
The skill is licensed under MIT, as stated on its page. Check the license file in the download for the full terms.
Does the skill send my data over the network?
No. The skill is listed as not using the network. It runs local Python scripts, so your product data stays on your machine.
Can the skill replace customer research?
No. The source says the scores and frameworks are generic and are not a substitute for real customer research. Use the outputs to structure your thinking, then verify with users.
Is cpo-advisor good for feature prioritisation?
Not for feature-level or day-to-day ticket work, which the skill explicitly excludes. It is designed for portfolio-level and strategic prioritisation.
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 cpo-advisor if you are a founder or product leader at a scaling company who needs a structured way to think about portfolio priorities, PMF, product org design or board reporting. Pick something else if you need feature-level or day-to-day ticket help, or if you want a verdict from scores alone, since the scripts use simplified weights and sample thresholds.
Download the skill