Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Converge all accumulated context into a crystallized North Star statement and structured ResearchBrief. Performs self-review before presenting to user. Use as the final tactic in any start mode — this is where everything comes together.
.claude/skills/yogsoth-ai-north-star-synthesis/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-02 | ✗→✓ | ▲ Improved | -6% | 0% |
| case-04 | ✗→✓ | ▲ Improved | -14% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-09 | ✗→✓ | ▲ Improved | -25% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 9% | 0% |
Converge all accumulated context into North Star + ResearchBrief.
| SOP | Purpose | Execution | |-----|---------|-----------| | crystallize-north-star | Formulate the North Star statement with user | dialogue | | generate-research-brief | Aggregate all context into structured ResearchBrief | subagent | | final-validation | Self-review + present to user for confirmation | dialogue |
crystallize-north-star checks: specific? ambitious? achievable?generate-research-brief aggregates all context (ActorProfile, FieldPanorama, RankedCandidates, ObstacleReport, GoalTree, North Star, key references, key terms)final-validation performs self-review before presenting:North Star (one sentence) + ResearchBrief (structured context document)
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | crystallize-north-star | Fuse the GoalTree root node and user motivation into a single crystallized North Star statement. Format: 'verb] specific goal], through method/path], solving what problem], ultimately what impact]'. Quality checks: specific? ambitious? achievable? | | final-validation | Self-review the North Star + ResearchBrief for completeness, consistency, and clarity before presenting to user. If issues found, return to specific tactic/SOP for targeted fix. If passes, present final output to user for confirmation. | | generate-research-brief | Aggregate all accumulated context from the crystallization process into a structured ResearchBrief document. This is the final output artifact alongside the North Star — a comprehensive requirement context document for downstream research strategies. |
<!-- 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→fail | 12,276 | 21,392 | +74% | 1 | 1 | 0% | 1,249 | 3,086 | +147% | 0 | 0 | — |
case-02 | fail→pass | 28,005 | 16,890 | -40% | 1 | 1 | 0% | 3,507 | 3,307 | -6% | 0 | 0 | — |
case-03 | fail→fail | 11,077 | 19,458 | +76% | 1 | 1 | 0% | 990 | 2,887 | +192% | 0 | 0 | — |
case-04 | fail→pass | 16,538 | 11,562 | -30% | 1 | 1 | 0% | 1,878 | 1,624 | -14% | 0 | 0 | — |
case-05 | fail→pass | 19,053 | 15,996 | -16% | 1 | 1 | 0% | 1,967 | 2,325 | +18% | 0 | 0 | — |
case-06 | pass→pass | 25,321 | 13,919 | -45% | 1 | 1 | 0% | 1,871 | 1,897 | +1% | 0 | 0 | — |
case-07 | fail→fail | 11,490 | 13,388 | +17% | 1 | 1 | 0% | 1,776 | 1,768 | -0% | 0 | 0 | — |
case-08 | pass→pass | 15,709 | 4,114 | -74% | 1 | 1 | 0% | 1,626 | 1,199 | -26% | 0 | 0 | — |
case-09 | fail→pass | 8,324 | 7,805 | -6% | 1 | 1 | 0% | 1,395 | 1,042 | -25% | 0 | 0 | — |
case-10 | fail→pass | 19,151 | 17,492 | -9% | 1 | 1 | 0% | 2,154 | 2,358 | +9% | 0 | 0 | — |
case-11 | pass→pass | 20,637 | 7,778 | -62% | 1 | 1 | 0% | 1,425 | 933 | -35% | 0 | 0 | — |
case-12 | fail→fail | 19,196 | 12,494 | -35% | 1 | 1 | 0% | 1,785 | 1,953 | +9% | 0 | 0 | — |
case-13 | pass→pass | 19,342 | 2,514 | -87% | 1 | 1 | 0% | 2,299 | 792 | -66% | 0 | 0 | — |
case-14 | pass→pass | 11,367 | 1,764 | -84% | 1 | 1 | 0% | 1,694 | 809 | -52% | 0 | 0 | — |
case-15 | fail→pass | 9,066 | 2,784 | -69% | 1 | 1 | 0% | 1,381 | 988 | -28% | 0 | 0 | — |
case-16 | fail→pass | 16,908 | 5,257 | -69% | 1 | 1 | 0% | 1,765 | 1,354 | -23% | 0 | 0 | — |
case-17 | pass→pass | 17,030 | 23,696 | +39% | 1 | 1 | 0% | 2,609 | 3,250 | +25% | 0 | 0 | — |
case-18 | pass→pass | 49,452 | 36,751 | -26% | 1 | 1 | 0% | 2,510 | 2,412 | -4% | 0 | 0 | — |
case-19 | pass→pass | 10,063 | 2,977 | -70% | 1 | 1 | 0% | 1,340 | 931 | -31% | 0 | 0 | — |
case-20 | fail→fail | 55,698 | 27,368 | -51% | 1 | 1 | 0% | 5,295 | 4,958 | -6% | 0 | 0 | — |
case-21 | fail→fail | 26,845 | 50,338 | +88% | 1 | 1 | 0% | 4,652 | 5,669 | +22% | 0 | 0 | — |
case-22 | pass→fail | 12,986 | 8,937 | -31% | 1 | 1 | 0% | 2,356 | 2,113 | -10% | 0 | 0 | — |
case-23 | fail→fail | 8,434 | 7,637 | -9% | 1 | 1 | 0% | 1,302 | 1,774 | +36% | 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. 23 cases were attempted. The headline lift of +26 percentage points is the difference between those two pass rates over the 23 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.