Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →levnikolaevich/ln-51-opportunity-evaluator
Evaluates new product directions using current demand, acquisition, competition, economics, and validation evidence. Use before commitment; not for backlog or implementation planning.
Passed 20 of 21 test cases with the skill, and 11 of 21 without it.
and 12 more in the full report
10 fixed11 passed either way1 still failing0 broke
Tokens, turns and time describe resource use, not a dollar price.
One of 10 test cases the skill fixed. Same model, same prompt, with and without the skill.
We are looking to expand our developer platform and are considering three potential product ideas: 1) An automated schema migration advisor, 2) A synthetic test data generator, and 3) An API deprecation tracker. Please evaluate these opportunities to help us decide which option to validate first. Please structure your analysis into five key sections: 1. Result: State the core verdict and primary outcome. 2. Scope: Define evaluated scope, exclusions, baseline metrics, and foundational assumptions. 3. Evidence: Separate our studio thesis from external market evidence. Compare candidates on demand signals, customer acquisition channels, competitive wedges, economic factors, and validation costs. Disclose source dates, confidence levels, and evidence classes. Provide decisive rejection reasons for eliminated or deferred candidates, and detail a low-cost validation experiment plan with budget boundaries and stop rules. 4. Verification: Detail verification checks performed, missing market signals, and state status. 5. Completion: Report checklist progress (X/Y completed), explanations for any unproven items with next actions, residual risks, and required decisions.
| Checked | Without skill | With skill |
|---|---|---|
| Uses one of the standard outcome tokens RECOMMEND, INCONCLUSIVE, or BLOCKED in the Result section. | Missed | Passed |
| Classifies market data using standard evidence categories like MEASURED, REPORTED, ESTIMATED, INFERRED, or UNKNOWN. | Missed | Passed |
The instructions your model reads when the skill loads.
Goal: Evaluate product opportunities before implementation commitment. Start from observable demand and a reachable acquisition path, eliminate weak candidates early, and recommend one low-cost validation step without manufacturing market precision.
Execution contract: The ordered checkboxes are the Definition of Done. Track every item internally as PENDING, PROVEN with concrete evidence, CLEARED with evidence that its condition is absent, or UNPROVEN with a gap; reading, delegation, or tool failure is not proof. Reconcile items after each section. Before returning, resolve all PENDING and count only PROVEN and CLEARED; apply the skill's verdict and approval rules to every gap.
Preserve user intent, scope, and existing authorization. Continue authorized work; ask only for consequential unresolved choices or required external approval. Scale depth to material risk without silently skipping checks. Preserve dependency and safety ordering; otherwise choose the verification method appropriate to each obligation.
| Need | Preferred tool | Use it when | Fallback |
|---|---|---|---|
| Product and constraints | User context plus existing product, analytics, customer, and strategy documents | Establishing audience, assets, channels, economics, and non-goals | State assumptions and request only consequential missing intent |
| Current demand and acquisition | Web research, trend or marketplace data, communities, reviews, ads, directories, and primary customer evidence | Every external market claim that affects elimination or recommendation | Mark the signal unavailable; never infer a number from search-result count |
| Competition and pricing | Competitor product pages, pricing, release history, distribution channels, reviews, and public filings where relevant | Establishing substitutes, willingness-to-pay signals, and credible differentiation | Use qualitative evidence with explicit confidence |
| Feasibility and validation cost | Existing capabilities, public APIs, regulations, platform rules, and current official documentation | Comparing the cheapest credible experiment and major blockers | Label estimates and name the evidence still required |
Keep the evaluation read-only. Do not create project files, roadmaps, Epics, Stories, implementation plans, campaigns, listings, advertisements, or customer outreach.
| Class | Meaning |
|---|---|
MEASURED | Direct analytics, transactions, experiments, or instrumented observations with known method and date |
REPORTED | A primary source reports a value or behavior, but the underlying measurement is not independently available |
ESTIMATED | A stated model based on explicit inputs and assumptions |
INFERRED | A qualitative conclusion from observable proxies |
UNKNOWN | Evidence is unavailable, stale, incomparable, or too weak to support a decision |
Do not turn REPORTED, ESTIMATED, or INFERRED evidence into a measured market size, search volume, conversion rate, revenue, or willingness-to-pay claim. Date every external source and distinguish the event date from the publication date when they differ.
Treat the creator thesis, intended experience, taste, and conviction as owner preferences and strategic-fit inputs, never as demand, acquisition, or willingness-to-pay evidence.
test cases passed, vs 11 / 21 without
More skills from the same source and nearby in the registry, with their headline benchmark lift.