Use this skill when approaching complex or ambiguous problems, breaking down large challenges, applying first principles thinking, or structuring your approach to a new problem. Trigger on keywords: how do I approach, I don't know where to start, complex problem, break this down, first principles, problem decomposition, stuck on, overwhelmed by.
SkillScore86Maintained
Automatic
◆Coding pack
Last verified 8/3/2026
What it changes
Passed 19 of 22 test cases with the skill, and 11 of 22 without it.
Without skill50%
With skill86%
Checks it passed that the model alone missed
The response specifies a 30-minute maximum limit.
The response includes an explicit 'Revisit if:' field.
The response identifies 'CONSTRAIN' as the second step.
and 5 more in the full report
8 fixed11 passed either way3 still failing0 broke
Pass-rate lift
+36 pp
Tokens
+12%
Agent turns
0%
Response time
-15%
Model gemini-3.6-flashMeasured 8/3/2026v119 of 22 cases passed with skill
Tokens, turns and time describe resource use, not a dollar price.
See it in action
One of 8 test cases the skill fixed. Same model, same prompt, with and without the skill.
1 of 8
Prompt
Our team has spent hours arguing over how to speed up database migrations. We are completely blocked. What is the very first action recommended when stuck on a hard problem?
What was checked, without and with the skill
Checked
Without skill
With skill
The response recommends restating the problem in different words.
Missed
Passed
case-21 · gemini-3.6-flash · 8/3/2026 · answers are capped and redacted
What's inside
The instructions your model reads when the skill loads.
SKILL.md92 linessingle file
Problem Solving
Core Mindset
Don't solve the problem you think you have. Solve the problem you actually have.
Most wasted effort in software comes from solving the wrong problem confidently. The first step is always to verify the problem definition.
The Problem-Solving Framework
1. DEFINE — What is the actual problem? (not the symptom)
2. CONSTRAIN — What are the boundaries? (must-haves, non-negotiables)
3. DECOMPOSE — Break into smaller sub-problems
4. EXPLORE — Generate multiple approaches before committing
5. DECIDE — Choose based on explicit criteria
6. EXECUTE — Implement the chosen approach
7. REVIEW — Did it solve the right problem?
First Principles Thinking
Instead of reasoning by analogy ("how do others do this?"), break the problem down to fundamental truths:
1. What do we know for certain is true about this problem?
2. What assumptions am I making that might be wrong?
3. If I had to rebuild this from scratch knowing only those truths,
what would I build?
Use when: existing solutions feel wrong, you're stuck in conventional thinking, or you need a breakthrough.