Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Use when complex problems require systematic step-by-step reasoning with ability to revise thoughts, branch into alternative approaches, or dynamically adjust scope. Ideal for multi-stage analysis, design planning, problem decomposition, or tasks with initially unclear scope.
.claude/skills/microck-sequential-thinking/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-08 | ✗→✓ | ▲ Improved | 18% | 0% |
| case-09 | ✗→✓ | ▲ Improved | 95% | 0% |
| case-10 | ✗→✓ | ▲ Improved | 68% | 0% |
| case-04 | ✓→✓ | = Same ✓ | 559% | 0% |
| case-05 | ✓→✓ | = Same ✓ | 322% | 0% |
Enables structured problem-solving through iterative reasoning with revision and branching capabilities.
Use mcp__reasoning__sequentialthinking when:
Don't use for: Simple queries, direct facts, or single-step tasks.
The MCP tool mcp__reasoning__sequentialthinking accepts these parameters:
thought (string): Current reasoning stepnextThoughtNeeded (boolean): Whether more reasoning is neededthoughtNumber (integer): Current step number (starts at 1)totalThoughts (integer): Estimated total steps neededisRevision (boolean): Indicates this revises previous thinkingrevisesThought (integer): Which thought number is being reconsideredbranchFromThought (integer): Thought number to branch frombranchId (string): Identifier for this reasoning branch1. Start with initial thought (thoughtNumber: 1)
2. For each step:
- Express current reasoning in `thought`
- Estimate remaining work via `totalThoughts` (adjust dynamically)
- Set `nextThoughtNeeded: true` to continue
3. When reaching conclusion, set `nextThoughtNeeded: false`typescript// First thought { thought: "Problem involves optimizing database queries. Need to identify bottlenecks first.", thoughtNumber: 1, totalThoughts: 5, nextThoughtNeeded: true } // Second thought { thought: "Analyzing query patterns reveals N+1 problem in user fetches.", thoughtNumber: 2, totalThoughts: 6, // Adjusted scope nextThoughtNeeded: true } // ... continue until done
For revision patterns, branching strategies, and complex workflows, see:
totalThoughts, refine as you progress| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-01 | fail→fail | 32,070 | 10,738 | -67% | 1 | 1 | 0% | 4,912 | 1,168 | -76% | 0 | 0 | — |
case-02 | fail→fail | 22,342 | 7,879 | -65% | 1 | 1 | 0% | 3,412 | 857 | -75% | 0 | 0 | — |
case-03 | fail→fail | 28,543 | 12,263 | -57% | 1 | 1 | 0% | 4,637 | 1,075 | -77% | 0 | 0 | — |
case-04 | pass→pass | 1,137 | 1,208 | +6% | 1 | 1 | 0% | 114 | 751 | +559% | 0 | 0 | — |
case-05 | pass→pass | 1,672 | 1,270 | -24% | 1 | 1 | 0% | 173 | 730 | +322% | 0 | 0 | — |
case-06 | pass→pass | 2,721 | 2,785 | +2% | 1 | 1 | 0% | 383 | 1,015 | +165% | 0 | 0 | — |
case-07 | fail→fail | 14,762 | 10,114 | -31% | 1 | 1 | 0% | 2,268 | 1,067 | -53% | 0 | 0 | — |
case-08 | fail→pass | 7,109 | 4,681 | -34% | 1 | 1 | 0% | 1,173 | 1,380 | +18% | 0 | 0 | — |
case-09 | fail→pass | 5,691 | 5,891 | +4% | 1 | 1 | 0% | 826 | 1,608 | +95% | 0 | 0 | — |
case-10 | fail→pass | 4,933 | 7,481 | +52% | 1 | 1 | 0% | 831 | 1,394 | +68% | 0 | 0 | — |
case-11 | pass→pass | 10,295 | 3,908 | -62% | 1 | 1 | 0% | 1,669 | 1,231 | -26% | 0 | 0 | — |
case-12 | fail→fail | 12,333 | 18,774 | +52% | 1 | 1 | 0% | 1,947 | 1,327 | -32% | 0 | 0 | — |
case-13 | fail→fail | 21,774 | 8,150 | -63% | 1 | 1 | 0% | 3,582 | 1,047 | -71% | 0 | 0 | — |
case-14 | fail→fail | 32,232 | 21,123 | -34% | 1 | 1 | 0% | 3,755 | 1,484 | -60% | 0 | 0 | — |
case-15 | fail→fail | 19,713 | 9,724 | -51% | 1 | 1 | 0% | 3,176 | 1,059 | -67% | 0 | 0 | — |
case-16 | fail→fail | 23,701 | 15,898 | -33% | 1 | 1 | 0% | 3,761 | 1,258 | -67% | 0 | 0 | — |
case-17 | fail→fail | 21,221 | 12,284 | -42% | 1 | 1 | 0% | 3,863 | 1,195 | -69% | 0 | 0 | — |
case-18 | fail→fail | 18,373 | 5,969 | -68% | 1 | 1 | 0% | 2,958 | 845 | -71% | 0 | 0 | — |
case-19 | fail→fail | 22,528 | 9,770 | -57% | 1 | 1 | 0% | 3,844 | 1,056 | -73% | 0 | 0 | — |
case-20 | fail→fail | 28,437 | 13,014 | -54% | 1 | 1 | 0% | 5,330 | 1,627 | -69% | 0 | 0 | — |
case-21 | fail→fail | 24,796 | 12,687 | -49% | 1 | 1 | 0% | 3,714 | 1,379 | -63% | 0 | 0 | — |
case-22 | fail→fail | 24,400 | 8,807 | -64% | 1 | 1 | 0% | 4,077 | 1,211 | -70% | 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, and 7 counted toward the lift figure. The other 15 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 +14 percentage points is the difference between those two pass rates over the 7 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.