Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Insight Campaign — deep root-cause analysis of why research gaps persist. 5 strategies (root-cause-drilling, stakeholder-mapping, tension-mining, question-reformulation, assumption-audit), 4 tactics, 13 subagent SOPs.
.claude/skills/yogsoth-ai-insight/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | 82% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 102% | 0% |
| case-18 | ✗→✓ | ▲ Improved | 61% | 0% |
| case-19 | ✗→✓ | ▲ Improved | 92% | 0% |
| case-23 | ✗→✓ | ▲ Improved | 42% | 0% |
Understand WHY gaps persist — root causes, stakeholders, tensions, and hidden assumptions.
This campaign is a strategy book — CC reads, internalizes, and autonomously constructs an approach. The SKILL.md files are textbooks, not scripts. CC decides execution order, depth, and iteration based on research context.
| Signal | Strategy | |--------|----------| | why it exists, root cause, 5 Whys, fishbone diagram, causal tree | → root-cause-drilling | | who is affected, stakeholders, JTBD, CSH 12 questions | → stakeholder-mapping | | opposing forces, tensions, polarities, evaporating cloud | → tension-mining | | how to frame the problem, HMW, abstraction ladder, Socratic | → question-reformulation | | hidden assumptions, fragility, premise audit, red team | → assumption-audit |
Import (5): web-search, web-research, paper-overview, paper-search, paper-research
Subagent (13): five-whys-drilling, ishikawa-decomposition, current-reality-tree, csh-12-question, jtbd-mapping, salience-classification, evaporating-cloud, polarity-mapping, abstraction-laddering, hmw-formulation, socratic-probing, abp-vulnerability-classification, clr-validation
Shared (2): evidence-synthesis, assumption-surfacing
context-checkpoint after each strategy completes. Accumulated state persists across strategies within a campaign run.
<!-- BEGIN available-tables (generated) -->
Optional, no fixed order; the final leaf is always a sop.
| Strategy | When to use | | --- | --- | | assumption-audit | Surface all assumptions, classify by vulnerability (load-bearing × likely-false), validate causal logic. Focus on dangerous assumptions — high load-bearing + non-explicit. | | question-reformulation | Reframe research questions using abstraction laddering, HMW formulation, and Socratic probing. Find the most productive level and framing for investigation. | | root-cause-drilling | Drill from surface symptoms to root causes via 5 Whys, Ishikawa decomposition, and Current Reality Trees. Validates each causal link with literature evidence. | | stakeholder-mapping | Map all affected parties using CSH 12-question framework, identify jobs-to-be-done, classify by salience. Reveals whose perspective is systematically excluded. | | tension-mining | Identify opposing forces that keep gaps open. Uses evaporating cloud to expose hidden assumptions behind conflicts and polarity mapping for unresolvable tensions. |
Optional, no fixed order; the final leaf is always a sop.
| SOP | When to use | | --- | --- | | context-checkpoint | Append research process and results to the current Phase's context file. Covers both process and results with genuine substance. Use this skill at plan-designated checkpoint points — typically after each strategy completes or at key decision nodes within a research Phase. | | context-init | Create a new context file for a research Phase. Called once at Phase start to initialize the file that subsequent context-checkpoint calls will append to. Use this skill whenever a new research Phase begins and a fresh context file is needed. |
<!-- 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 | 65,082 | 76,277 | +17% | 1 | 1 | 0% | 5,572 | 8,010 | +44% | 0 | 0 | — |
case-02 | fail→fail | 33,554 | 89,752 | +167% | 1 | 1 | 0% | 3,883 | 1,525 | -61% | 0 | 0 | — |
case-03 | fail→pass | 30,337 | 79,745 | +163% | 1 | 1 | 0% | 3,193 | 5,799 | +82% | 0 | 0 | — |
case-04 | fail→fail | 23,064 | 28,405 | +23% | 1 | 1 | 0% | 3,167 | 1,618 | -49% | 0 | 0 | — |
case-05 | pass→pass | 18,604 | 27,086 | +46% | 1 | 1 | 0% | 3,052 | 5,437 | +78% | 0 | 0 | — |
case-06 | pass→pass | 26,067 | 37,337 | +43% | 1 | 1 | 0% | 3,687 | 7,214 | +96% | 0 | 0 | — |
case-07 | pass→fail | 52,234 | 21,808 | -58% | 1 | 1 | 0% | 5,028 | 1,518 | -70% | 0 | 0 | — |
case-08 | fail→fail | 24,604 | 42,893 | +74% | 1 | 1 | 0% | 3,956 | 1,679 | -58% | 0 | 0 | — |
case-09 | pass→pass | 16,010 | 30,078 | +88% | 1 | 1 | 0% | 2,496 | 4,832 | +94% | 0 | 0 | — |
case-10 | fail→pass | 30,262 | 44,982 | +49% | 1 | 1 | 0% | 4,047 | 8,173 | +102% | 0 | 0 | — |
case-11 | pass→pass | 48,043 | 38,089 | -21% | 1 | 1 | 0% | 8,239 | 9,032 | +10% | 0 | 0 | — |
case-12 | pass→pass | 39,497 | 38,479 | -3% | 1 | 1 | 0% | 6,378 | 9,032 | +42% | 0 | 0 | — |
case-13 | pass→pass | 37,512 | 39,343 | +5% | 1 | 1 | 0% | 7,984 | 9,023 | +13% | 0 | 0 | — |
case-14 | pass→pass | 31,381 | 39,693 | +26% | 1 | 1 | 0% | 4,732 | 6,810 | +44% | 0 | 0 | — |
case-15 | pass→pass | 22,353 | 44,729 | +100% | 1 | 1 | 0% | 3,614 | 6,294 | +74% | 0 | 0 | — |
case-16 | pass→pass | 28,317 | 49,030 | +73% | 1 | 1 | 0% | 5,026 | 9,033 | +80% | 0 | 0 | — |
case-17 | pass→pass | 14,733 | 29,069 | +97% | 1 | 1 | 0% | 2,353 | 5,482 | +133% | 0 | 0 | — |
case-18 | fail→pass | 22,235 | 42,242 | +90% | 1 | 1 | 0% | 3,353 | 5,399 | +61% | 0 | 0 | — |
case-19 | fail→pass | 19,858 | 33,483 | +69% | 1 | 1 | 0% | 3,179 | 6,109 | +92% | 0 | 0 | — |
case-20 | fail→fail | 17,546 | 20,306 | +16% | 1 | 1 | 0% | 2,822 | 1,752 | -38% | 0 | 0 | — |
case-21 | pass→pass | 21,557 | 51,394 | +138% | 1 | 1 | 0% | 3,328 | 9,026 | +171% | 0 | 0 | — |
case-22 | pass→pass | 16,651 | 29,791 | +79% | 1 | 1 | 0% | 2,448 | 5,131 | +110% | 0 | 0 | — |
case-23 | fail→pass | 24,508 | 24,929 | +2% | 1 | 1 | 0% | 3,477 | 4,941 | +42% | 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, and 18 counted toward the lift figure. The other 5 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +17 percentage points is the difference between those two pass rates over the 18 comparable cases. 3 cases got worse with the skill loaded, and they are 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.