Menu
Systematic debugging using hypothesis-driven investigation with test, narrow, and resolve loops
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/InugamiDev/ultrathink-oss --skill debugThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Showing SKILL.md
Unified design foundations — design system architecture, tokens, component specs, visual principles, creative vision, figma integration, plus brand design system loader (66 real brands via DESIGN.md). Absorbs design, design-system, design-systems, design-principles, design-router, creative-vision, figma, design-md.
Render, summarize, and present markdown documents and structured content in multiple output modes
Ultra UI skill - combines Google's DESIGN.md spec (machine-readable design tokens) with the ui-ux-pro-max knowledge base (91 styles, 161 palettes, 73 font pairings, 161 products, 104 UX guidelines, 25 chart types). Generates lint-clean DESIGN.md files, validates token references and WCAG contrast, exports Tailwind/DTCG tokens, and diffs design systems version-over-version.
Based on SOC occupation classification
| name | debug |
| description | Systematic debugging using hypothesis-driven investigation with test, narrow, and resolve loops |
| layer | hub |
| category | workflow |
| triggers | ["/debug","debug this","why is this broken","find the bug","this is not working","something is wrong with"] |
| inputs | [{"symptom":"Description of the observed incorrect behavior"},{"expected":"What the correct behavior should be (optional but helpful)"},{"context":"Error messages, stack traces, reproduction steps, recent changes (optional)"}] |
| outputs | [{"rootCause":"Identified root cause with evidence"},{"hypothesis log":"Record of hypotheses tested and their outcomes"},{"fix recommendation":"Suggested fix (handed off to fix skill for application)"}] |
| linksTo | ["fix","scout","test","code-review"] |
| linkedFrom | ["cook","team","ship"] |
| preferredNextSkills | ["fix","test"] |
| fallbackSkills | ["scout","research"] |
| riskLevel | low |
| memoryReadPolicy | selective |
| memoryWritePolicy | selective |
| sideEffects | ["May run test commands to reproduce issues","May read many files during investigation","Produces hypothesis log in working memory"] |
Systematically find the root cause of bugs using a disciplined hypothesis-test-narrow loop. This skill treats debugging as a scientific investigation, not a guessing game.
The core principle: never change code to debug. Observe, hypothesize, test, narrow. Only hand off to fix once the root cause is identified with evidence.
Capture the symptom precisely
Gather initial evidence
Classify the bug type (this guides the investigation strategy)
Generate 2-5 hypotheses ranked by likelihood. For each hypothesis:
Prioritize hypotheses using these heuristics:
Test the highest-priority hypothesis
Record the result
Narrow the search space
Repeat from step 6 until root cause is identified or all hypotheses are exhausted.
State the root cause clearly
Verify with evidence
If all hypotheses exhausted without finding root cause:
scout for broader codebase explorationresearch for known issues with the technologyfix skill for implementation.# Debug Report
## Symptom
**Observed**: [exact behavior]
**Expected**: [correct behavior]
**Reproducibility**: Always | Sometimes | Rare
**First noticed**: [when/context]
## Root Cause
**Location**: [file:line]
**Cause**: [precise description]
**Mechanism**: [how the cause produces the symptom]
## Evidence Chain
1. [observation 1] → indicates [conclusion 1]
2. [observation 2] → confirms [conclusion 2]
3. Therefore: [root cause]
## Hypothesis Log
| # | Hypothesis | Test | Result |
|---|-----------|------|--------|
| H1 | [statement] | [what was checked] | Eliminated: [reason] |
| H2 | [statement] | [what was checked] | Confirmed |
## Recommended Fix
[Description of what should change and why]
**Affected files**: [list]
**Risk**: Low | Medium | High
## Regression Prevention
[What test should be written to catch this if it recurs]
/debug TypeError: Cannot read property 'map' of undefined in UserList component
/debug The checkout flow completes but the order is not saved to the database
/debug Login stopped working after merging PR #42. Users get a 401 even with valid credentials.
/debug We eliminated the auth middleware hypothesis. What else could cause 401s on the /api/user endpoint?
Symptom: "TypeError: Cannot read property 'map' of undefined" Hypotheses:
Investigation: Read the component, find it expects data.items.map(...). Check the API endpoint -- it returns { items: [...] } on success but { error: "..." } on failure with no items key.
Root cause: Missing error handling -- component assumes API always returns items.
Symptom: "Orders created via API are missing the discount field" Hypotheses:
Investigation: Check API route -- discount is in the request. Check ORM model -- discount field exists. Check database -- column exists. Check the create function -- found it: the createOrder function destructures the input but does not include discount in the destructured fields.
Root cause: Destructuring omission in createOrder() at src/lib/orders.ts:42.