agentsclimarketplace

Problem discovery

Skill ajstars1/startup-coach/problem-discovery

Agent-based startup coaching skills for Claude Code — every claim labeled FACT/ASSUMPTION/UNKNOWN. Zero-hallucination protocol, 10 skills, one master orchestrator.

Install
npx -y skills add ajstars1/startup-coach --skill problem-discovery

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

  • 0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Understand WHO has the problem and their world before designing solutions — Customer Persona, Problem Statement (mad-lib format), Empathy Map (Dave Gray), and Customer Ecosystem & Workflow mapping. Use early in validation, whenever the target customer is fuzzy, or when a founder describes their solution but can't crisply describe the person suffering the problem.

SKILL.md

4.6 KB, as published. Nobody here has run it

Problem Discovery — Persona, Problem Statement, Empathy Map, Ecosystem & Workflow

Four tools, run together (they feed each other), all producing labeled hypothesis drafts to be destroyed or confirmed by real interviews. Everything drafted pre-interview is [ASSUMPTION] by definition.

1. Customer Persona (provisional, pre-discovery)

Consider 2+ candidate archetypes first; pick a primary with justification, keep the runner-up listed. Then fill:

  1. Name (fictional, for role-playing — label it so)
  2. Description — demographics, goals, needs
  3. Key problems/needs your idea aims to solve
  4. Current solutions available to them (real, findable ones)
  5. Desired outcomes — what success looks like for them (must eventually come from their mouth, not yours)

Rebuild the persona from real quotes after interviews. Watch for the early-adopter tell: the person with the most acute version of the problem (per Talking to Humans) often isn't the "average" customer — real interviews frequently reveal a sharper archetype than the founder's initial guess.

2. Problem Statement — use this exact mad-lib

When <context> occurs, <customer> struggles with <problem>. Because of this, they feel <emotional response> and experience <quantifiable/practical impact>. Currently, they use <alternative solutions> despite <alternative shortcomings>.

Quality bars:

  • <quantifiable impact> must be a number with a source, or written as [UNKNOWN — measure via interviews]. Never invent it.
  • <alternatives> must be real. "We have no competitors" is almost always false — customers are doing something today (a shop visit, a spreadsheet, a workaround, postponing). The status quo is the real competitor.
  • Add: who else is affected if this continues, and who will pay for a solution (user ≠ buyer, often).
  • Honest check: if you have negative-space evidence (a silent launch, low signups), state what it does and doesn't imply — "no felt pain" vs "offer never reached them" are different diagnoses.

3. Empathy Map (Dave Gray, XPLANE — gamestorming.com)

Quadrants: Says / Thinks / Does / Feels, plus Pains and Gains. Pre-interview: fill "Says" with things to listen for — never invented quotes. Post-interview: replace with verbatim customer language. Rank pains/gains WITH customers, not for them.

4. Ecosystem Map + Customer Workflow

  • Ecosystem map: every entity around the customer — users, buyers, influencers, saboteurs, suppliers, channels, regulators, payment methods. Mark the critical roles in the purchase, consumption, and payment decisions. Identify the possible saboteur (whoever loses if you win — often the incumbent relationship).
  • Customer workflow: the step-by-step process the customer follows TODAY, without your product — where time, money, and frustration occur, and where your solution would slot in. Draw today's workflow, not the imagined future one.
  • The bundling check: friction that looks separable often isn't (e.g., if two errands always happen in one trip, removing one saves less than it appears). Find which steps are truly independent before valuing their removal.
  • Quantify frictions as [ASSUMPTION → measure]: steps, trips, minutes, delay rates, rework rates.

Output

Persona + problem statement + empathy map + ecosystem/workflow (mermaid diagrams welcome), every field labeled [FACT/ASSUMPTION/UNKNOWN], ending with the 5 highest-value interview questions this analysis raises — the goal of these drafts is to be proven wrong cheaply. Hand those questions to /customer-discovery.

Credits: Empathy Map — Dave Gray (XPLANE). Persona/problem-statement/workflow structures as practiced in evidence-first incubation programs; customer archetype discipline — Steve Blank.

Ledger discipline (v2): reference Belief Ledger IDs ([A4], [F7]) for every claim; propose new/changed items in a closing "Ledger delta" section rather than restating claims — restated claims drift.

Fill-in templates: TEMPLATES.md in this skill's folder — copy the structures into your project's discovery/ and fill with labels.

Keep looking

Skills are one crate of 328,083. Ordering is by how many stacks a row turns up in, so the top of any crate is what has actually been picked rather than what has the most stars.