Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Estimate resources, budget, and timeline using parametric, analogous, and three-point (PERT) estimation methods.
.claude/skills/yogsoth-ai-resource-envelope-estimation/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -16% | 0% |
| case-03 | ✗→✓ | ▲ Improved | -29% | 0% |
| case-06 | ✗→✓ | ▲ Improved | 60% | 0% |
| case-08 | ✗→✓ | ▲ Improved | 23% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -26% | 0% |
Purpose: Produce credible resource estimates for implementing a candidate. Combines parametric estimation (model-based), analogous estimation (reference-class), and three-point PERT estimation (optimistic/likely/pessimistic) to bound the resource envelope with quantified uncertainty.
When to use:
| Metric | Target | |--------|--------| | Estimate dimensions | >= 3 (time, cost, personnel) | | Precision range | +/-30% initial, +/-10% refined | | Reference analogies | >= 2 per estimate |
| Key | Type | Description | |-----|------|-------------| | candidate | object | The candidate being estimated | | parametric_estimate | object | Model-based estimate | | analogous_estimate | object | Reference-class estimate | | pert_estimate | object | Three-point estimate | | envelope | object | Synthesized resource envelope | | confidence_level | float | Confidence in the estimate |
| Tactic | When | |--------|------| | multi-dimensional-readiness-scan | When resource estimation requires understanding current maturity first | | staged-gate-evaluation | When resources need to be estimated per stage gate |
| SOP | Purpose | |-----|---------| | dimension-assessment | Assess resource dimension readiness | | gate-criteria-definition | Define resource gates | | feasibility-synthesis | Synthesize resource estimates into overall feasibility |
yamlresource_envelope: candidate: <name> time: optimistic: <duration> likely: <duration> pessimistic: <duration> expected: <PERT weighted> cost: optimistic: <amount> likely: <amount> pessimistic: <amount> expected: <PERT weighted> personnel: roles: [{role, count, duration}] total_person_months: N analogies_used: [{project, relevance, actual_cost, actual_time}] confidence: 0.X precision_band: "+/-N%" key_assumptions: [...]
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Tactic | When to use | | --- | --- | | multi-dimensional-readiness-scan | Assess readiness across multiple dimensions, synthesize into radar visualization, and identify bottleneck dimensions. | | staged-gate-evaluation | Define gate criteria for each stage, evaluate candidates at each gate, and render go/kill/recycle decisions with evidence. |
<!-- END available-tables (generated) -->
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→pass | 30,515 | 21,489 | -30% | 1 | 1 | 0% | 5,109 | 4,271 | -16% | 0 | 0 | — |
case-02 | fail→fail | 38,483 | 17,983 | -53% | 1 | 1 | 0% | 7,529 | 3,369 | -55% | 0 | 0 | — |
case-03 | fail→pass | 37,387 | 20,436 | -45% | 1 | 1 | 0% | 6,660 | 4,761 | -29% | 0 | 0 | — |
case-04 | pass→pass | 4,629 | 17,431 | +277% | 1 | 1 | 0% | 717 | 2,581 | +260% | 0 | 0 | — |
case-05 | fail→fail | 11,827 | 13,989 | +18% | 1 | 1 | 0% | 747 | 2,209 | +196% | 0 | 0 | — |
case-06 | fail→pass | 16,245 | 22,876 | +41% | 1 | 1 | 0% | 2,069 | 3,310 | +60% | 0 | 0 | — |
case-07 | pass→pass | 27,644 | 16,772 | -39% | 1 | 1 | 0% | 3,959 | 3,126 | -21% | 0 | 0 | — |
case-08 | fail→pass | 28,236 | 17,124 | -39% | 1 | 1 | 0% | 2,521 | 3,091 | +23% | 0 | 0 | — |
case-09 | fail→pass | 25,158 | 8,675 | -66% | 1 | 1 | 0% | 3,412 | 2,528 | -26% | 0 | 0 | — |
case-18 | fail→pass | 22,966 | 26,239 | +14% | 1 | 1 | 0% | 5,373 | 3,525 | -34% | 0 | 0 | — |
case-10 | pass→pass | 21,942 | 24,363 | +11% | 1 | 1 | 0% | 2,614 | 3,647 | +40% | 0 | 0 | — |
case-11 | pass→pass | 17,421 | 15,941 | -8% | 1 | 1 | 0% | 3,085 | 2,995 | -3% | 0 | 0 | — |
case-12 | fail→pass | 18,977 | 31,719 | +67% | 1 | 1 | 0% | 2,567 | 3,373 | +31% | 0 | 0 | — |
case-13 | pass→pass | 18,692 | 19,689 | +5% | 1 | 1 | 0% | 3,371 | 4,763 | +41% | 0 | 0 | — |
case-14 | fail→pass | 30,124 | 19,242 | -36% | 1 | 1 | 0% | 4,729 | 3,353 | -29% | 0 | 0 | — |
case-15 | pass→pass | 14,569 | 17,342 | +19% | 1 | 1 | 0% | 2,711 | 3,152 | +16% | 0 | 0 | — |
case-16 | pass→pass | 5,905 | 11,034 | +87% | 1 | 1 | 0% | 970 | 2,825 | +191% | 0 | 0 | — |
case-17 | pass→pass | 34,436 | 21,058 | -39% | 1 | 1 | 0% | 954 | 1,911 | +100% | 0 | 0 | — |
case-19 | fail→pass | 22,032 | 12,863 | -42% | 1 | 1 | 0% | 3,672 | 3,214 | -12% | 0 | 0 | — |
case-20 | pass→pass | 30,644 | 13,115 | -57% | 1 | 1 | 0% | 2,449 | 2,937 | +20% | 0 | 0 | — |
case-21 | pass→pass | 20,801 | 24,317 | +17% | 1 | 1 | 0% | 3,573 | 3,760 | +5% | 0 | 0 | — |
case-22 | pass→fail | 19,168 | 24,743 | +29% | 1 | 1 | 0% | 3,037 | 4,799 | +58% | 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 +36 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
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.