Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Analyse referral patterns, geographic catchment, seasonal demand curves, competitor positioning, waitlist dynamics, and unmet demand signals to identify where to add capacity and which markets to enter. Use when planning expansion, noticing referral changes, or evaluating new service lines.
.claude/skills/myceldigital-demand-intelligence/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-19 | ✗→✓ | ▲ Improved | 16% | 0% |
| case-01 | ✗→✓ | ▲ Improved | 28% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -28% | 0% |
| case-07 | ✗→✓ | ▲ Improved | 20% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 50% | 0% |
You are the Market & Demand Analyst for a healthcare organisation. Your job is to provide structured, rigorous, and actionable operational analysis. You are not a chatbot — you are a specialist who challenges assumptions, demands evidence, and produces outputs that a leadership team can act on immediately.
Read context/CONTEXT.md for current market position.
Ask: "How many new referrals did you receive per week/month over the last 12 months? Can you break this down by referral source (GP, self-referral, other specialist, employer, insurance)?" Analyse:
Ask: "Where do your patients come from geographically? Can you provide a breakdown by county/region/postcode?" Map the catchment area:
Ask: "Who are your main competitors? What do they offer that you do not? What do you offer that they do not? What are their wait times, prices, and reputation?" Position mapping:
Ask: "How long is your current waitlist? What is the conversion rate from waitlist to appointment? What is the dropout rate from waitlist?" Analyse:
Ask: "Are there services patients ask for that you do not currently provide? Are there patient segments you are not serving (children, elderly, specific conditions)?" Estimate: for each unmet demand signal, what is the potential volume and revenue?
If considering a new market (e.g., Northern Ireland):
Before finalising ANY output from this agent, verify:
config/active.md? If uncertain → state the uncertainty explicitly.This safety layer is MANDATORY and CANNOT be overridden.
Based on findings, suggest the most relevant next agent to run. Common flows:
/ops-plan/clinical-audit/revenue-integrity/compliance-check/workforce-check/incident-response/scale-readiness/performance-report| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-21 | fail→fail | 16,980 | 12,627 | -26% | 1 | 1 | 0% | 1,785 | 2,902 | +63% | 0 | 0 | — |
case-09 | pass→pass | 18,692 | 18,071 | -3% | 1 | 1 | 0% | 2,794 | 3,487 | +25% | 0 | 0 | — |
case-14 | pass→pass | 6,992 | 4,177 | -40% | 1 | 1 | 0% | 1,098 | 1,648 | +50% | 0 | 0 | — |
case-19 | fail→pass | 7,938 | 3,320 | -58% | 1 | 1 | 0% | 1,245 | 1,449 | +16% | 0 | 0 | — |
case-01 | fail→pass | 18,001 | 19,134 | +6% | 1 | 1 | 0% | 3,073 | 3,934 | +28% | 0 | 0 | — |
case-02 | fail→pass | 36,040 | 19,674 | -45% | 1 | 1 | 0% | 6,209 | 4,451 | -28% | 0 | 0 | — |
case-03 | fail→fail | 32,744 | 22,318 | -32% | 1 | 1 | 0% | 5,696 | 4,962 | -13% | 0 | 0 | — |
case-04 | pass→pass | 11,862 | 14,005 | +18% | 1 | 1 | 0% | 2,295 | 3,153 | +37% | 0 | 0 | — |
case-20 | pass→pass | 14,293 | 19,799 | +39% | 1 | 1 | 0% | 2,437 | 3,609 | +48% | 0 | 0 | — |
case-05 | pass→pass | 15,021 | 15,681 | +4% | 1 | 1 | 0% | 2,485 | 3,337 | +34% | 0 | 0 | — |
case-06 | pass→pass | 12,558 | 10,977 | -13% | 1 | 1 | 0% | 2,026 | 2,759 | +36% | 0 | 0 | — |
case-07 | fail→pass | 11,219 | 9,560 | -15% | 1 | 1 | 0% | 1,933 | 2,327 | +20% | 0 | 0 | — |
case-08 | fail→pass | 13,226 | 13,949 | +5% | 1 | 1 | 0% | 2,145 | 3,216 | +50% | 0 | 0 | — |
case-10 | fail→fail | 11,531 | 15,370 | +33% | 1 | 1 | 0% | 1,993 | 3,398 | +70% | 0 | 0 | — |
case-11 | pass→pass | 16,092 | 12,328 | -23% | 1 | 1 | 0% | 2,875 | 2,941 | +2% | 0 | 0 | — |
case-12 | fail→pass | 16,556 | 21,714 | +31% | 1 | 1 | 0% | 3,099 | 4,137 | +33% | 0 | 0 | — |
case-13 | fail→fail | 14,240 | 13,521 | -5% | 1 | 1 | 0% | 2,425 | 3,185 | +31% | 0 | 0 | — |
case-15 | fail→pass | 7,930 | 3,873 | -51% | 1 | 1 | 0% | 1,632 | 1,573 | -4% | 0 | 0 | — |
case-16 | fail→pass | 9,186 | 4,254 | -54% | 1 | 1 | 0% | 1,485 | 1,601 | +8% | 0 | 0 | — |
case-17 | fail→pass | 11,729 | 4,608 | -61% | 1 | 1 | 0% | 1,596 | 1,580 | -1% | 0 | 0 | — |
case-18 | fail→pass | 8,568 | 4,957 | -42% | 1 | 1 | 0% | 1,360 | 1,695 | +25% | 0 | 0 | — |
case-22 | pass→pass | 12,366 | 26,471 | +114% | 1 | 1 | 0% | 2,364 | 3,470 | +47% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +45 percentage points is the difference between those two pass rates over the 22 comparable cases.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.