sequential-thinking
Structured reflective problem-solving methodology. Process: decompose, analyze, hypothesize, verify, revise. Capabilities: complex problem decomposition, adaptive planning, course correction, hypothesis verification, multi-step analysis. Actions: decompose, analyze, plan, revise, verify solutions step-by-step. Keywords: sequential thinking, problem decomposition, multi-step analysis, hypothesis verification, adaptive planning, course correction, reflective thinking, step-by-step, thought sequence, dynamic adjustment, unclear scope, complex problem, structured analysis. Use when: decomposing complex problems, planning with revision capability, analyzing unclear scope, verifying hypotheses, needing course correction, solving multi-step problems.
79
Quality
83%
Does it follow best practices?
Impact
54%
3.17xAverage score across 3 eval scenarios
Passed
No known issues
Evaluation results
40%
35%Diagnosing a Mysteriously Failing API Integration Test
Hypothesis-driven debugging
Numbered thought format
0%
100%
Dynamic total adjustment
0%
0%
One aspect per thought
50%
75%
HYPOTHESIS label present
0%
0%
VERIFICATION label present
0%
0%
REVISION label present
0%
0%
Revision content complete
0%
0%
FINAL label present
0%
100%
Thought signals next
0%
100%
Context built explicitly
12%
0%
FINAL only after confidence
0%
75%
Without context: $0.3061 · 2m 10s · 11 turns · 18 in / 6,738 out tokens
With context: $0.3097 · 2m 2s · 11 turns · 12 in / 5,945 out tokens
61%
30%Choosing a Data Store for a Real-Time Leaderboard Service
Branching and convergence
Numbered thought format
0%
100%
BRANCH label present
0%
0%
Branches share same thought number
0%
0%
Explicit comparison
83%
100%
Convergence with rationale
100%
100%
FINAL label present
0%
100%
Dynamic total adjustment
0%
0%
Context built explicitly
25%
50%
Assumptions stated
77%
100%
Thought signals next
0%
100%
Without context: $0.2189 · 1m 28s · 8 turns · 13 in / 3,970 out tokens
With context: $0.5888 · 3m 18s · 23 turns · 278 in / 9,622 out tokens
62%
45%Planning a Legacy Authentication System Refactor
Adaptive planning with revision
Numbered thought format
0%
100%
Dynamic total adjustment
0%
0%
REVISION label present
0%
0%
Revision content complete
0%
0%
One aspect per thought
50%
100%
Assumptions or uncertainties stated
75%
100%
Thought signals next
0%
100%
Context built explicitly
25%
50%
FINAL label present
0%
100%
FINAL only when ready
0%
100%
Explicit mode used appropriately
62%
100%
Without context: $0.2122 · 1m 30s · 8 turns · 11 in / 4,252 out tokens
With context: $0.4184 · 2m 23s · 17 turns · 17 in / 6,899 out tokens
- Repository
- majiayu000/claude-skill-registry-data
- Commit
2bbaa03
- Evaluated
- Agent
- Claude Code
- Model
- Claude Sonnet 4.6
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.