Secondary research
Agent-based startup coaching skills for Claude Code — every claim labeled FACT/ASSUMPTION/UNKNOWN. Zero-hallucination protocol, 10 skills, one master orchestrator.
npx -y skills add ajstars1/startup-coach --skill secondary-researchAssembled 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
Structured desk research for startups — demand, competition, value chain, customer types, and purchase process — that converts assumptions into sourced facts (URL + publisher + date for every figure) and honestly reports what desk research cannot answer. Use for market sizing, competitor scans, regulatory questions, or whenever an F/A/U list has researchable unknowns.
SKILL.md
3.6 KB, as published. Nobody here has run it
Secondary Research
Purpose: move Assumptions → Facts where possible — before and alongside customer discovery, never instead of it. Not a one-off: understanding your business environment is continuous.
Where to start
Which industry/sector should you research? Hint: who are your likely competitors — which sector are they in? If you believe you have no competitor, think again: your target customers are using something right now (a product, a workaround, a shop visit, postponement). Whatever they use defines the industry to analyse — the status quo is the real competitor.
The five focus areas (checklist, not exhaustive)
a) Demand — demand pattern; unit of consumption; growth rate vs GDP; seasonality or concentration (festivals, fiscal years, weather); thousands of customers or a few; new vs established market; does one sale trigger follow-on revenue (printers→cartridges, cars→service)?
b) Competition — how many suppliers; basis of competition (price, quality, reliability, status, convenience); intensity; what determines market share; new entrants in the past 3 years and their entry strategies; who has tried this model and failed — and why (failed lookalikes are the cheapest education available).
c) Value chain — every entity between originator and consumer (distributor, transporter, regulator, financier, platform); the role of each; who plays the critical roles in the purchase, consumption, and payment decisions.
d) Customer types — how many distinct types exist; the differences between them.
e) Purchase process — when the need is felt; how they learn what/where to buy; how they decide among competitors; the actual steps; anything odd that stands out.
Evidence rules (non-negotiable)
- Every figure needs URL + publisher + date. Prefer: government/statistical data, filings, credible trade press, primary company sources. Note when data is old (e.g., a decade-old census) and label derived estimates as derived.
- Sources conflict → report both. Nothing credible found → the item stays UNKNOWN — say so plainly. Never fill a gap with a plausible number.
- Grade confidence per finding: HIGH (official/primary-adjacent) / MEDIUM (single secondary source) / UNKNOWN.
- Separate global from local: national market stats rarely answer "how many buyers in THIS town" — derive carefully or mark unknown.
Output
Findings table (Focus area | Finding | Source URL | Confidence) → 3–5 key conclusions about the business environment (each labeled) → "What this cannot tell you — needs interviews" — the honest list of questions only customer discovery answers (willingness to pay, actual behavior, trust barriers, why a past launch failed). Feed conclusions back into /facts-assumptions-unknowns as explicit list movements.
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.