Back to Skills

pol-probe-advisor

Select the right Proof of Life (PoL) probe based on hypothesis, risk, and resources. Use this to match the validation method to the real learning goal, not tooling comfort.

6,261stars760forksUpdated 8/5/2026

Security Assessment

Safe(98/100)
Security Score98/100

About pol-probe-advisor

This skill is an interactive advisor that helps product managers select the right Proof of Life (PoL) probe type based on their hypothesis, risk, and available resources. It solves the common method-hypothesis mismatch problem, where PMs choose a validation method based on tooling comfort (for example defaulting to Figma or code) rather than the real learning goal, and end up validating the wrong thing and missing the actual risk.

Rather than deciding whether to validate (it assumes you should), the skill is a decision framework for choosing how to validate most effectively. It presents the five probe flavors (feasibility check, task-focused test, narrative prototype, synthetic data simulation, and vibe-coded probe) with their core questions, best-fit situations, and timelines, and it uses adaptive questioning to gather context: the hypothesis, the specific risk to eliminate, the timeline, and available resources such as engineering capacity, design tools, no-code and AI tools, user access, and budget. It then identifies the core question the user is really trying to answer and recommends a probe type, guided by the golden rule of using the cheapest prototype that tells the harshest truth. It defers to a companion workshop-facilitation skill for its one-question-at-a-time interaction protocol, progress labels, and numbered response options, and it explicitly lists anti-patterns and when not to use it.

Target users are product managers, founders, and innovation teams facing a risky idea who need to pick the most efficient validation approach before building. It is a pure advisory and planning framework with no code execution or system access.

FAQ

What does this advisor actually decide?

It recommends which of the five PoL probe flavors best fits your situation; it does not decide whether you should validate (it assumes you should).

What information does it need from me?

Your hypothesis, the specific risk you want to eliminate, your timeline, and your available resources such as engineering capacity, design and no-code tools, user access, and budget.

What are the five probe flavors?

Feasibility check, task-focused test, narrative prototype, synthetic data simulation, and vibe-coded probe, each matched to a different core question and timeline.

When should I not use this skill?

When you don't yet have a hypothesis, when you're trying to impress executives rather than validate, when you already know the answer, or when you need to ship an MVP rather than run pre-MVP reconnaissance.

Does it run on its own or ask questions?

It is interactive and uses adaptive, one-question-at-a-time questioning, deferring to a companion workshop-facilitation skill for its interaction protocol and numbered response options.

Install pol-probe-advisor

Download and extract the skill files to your .claude/skills/ directory.

Quick Setup:

  1. Copy the skill folder to .claude/skills/
  2. Claude will automatically detect and use the skill