debug
/debug
Use when encountering a bug, test failure, unexpected behavior, or when the user asks for help debugging. Hypothesis-driven — reproduce, investigate, fix, verify. Standalone — works outside the plan/spec/build workflow.
- Author
- ratler ↗
- Repository
- dream-team ↗
- Licence
- MIT
- Last repository change
- 16.10.2026
Assessment and security
- Files in this package
- 1
- Scripts
- 0
- HTML comments
- 0
- URLs in SKILL.md
- 0
No signals detected. Repository changed on 16.10.2026, without references to retired models or APIs.
MIT, open: use and modify while retaining author attribution.
What it does and when to use it
Use when encountering a bug, test failure, unexpected behavior, or when the user asks for help debugging. Hypothesis-driven — reproduce, investigate, fix, verify. Standalone — works outside the plan/spec/build workflow.
Arguments
/debug <describe what's broken>
Original author description
Use when encountering a bug, test failure, unexpected behavior, or when the user asks for help debugging. Hypothesis-driven — reproduce, investigate, fix, verify. Standalone — works outside the plan/spec/build workflow.
How it works
- The Process
- Report
- Key Principles
- When to Stop and Escalate
SKILL.md preview
--- name: debug description: "Use when encountering a bug, test failure, unexpected behavior, or when the user asks for help debugging. Hypothesis-driven — reproduce, investigate, fix, verify. Standalone — works outside the plan/spec/build workflow." argument-hint: "<describe what's broken>" --- # Debugging Through Investigation **If no arguments were provided, stop immediately and tell the user: "Usage: /dream-team:debug <describe what's broken>". Do not proceed.** Help the user debug an issue through systematic investigation and hypothesis-driven problem solving. This is a conversation, not a fix-it script. Start by understanding the issue, then reproduce it, form theories, investigate with evidence, and only fix once you understand the root cause. ## The Process **Understanding the issue:** - Explore the codebase first — read relevant files to build context before asking questions - Ask questions one at a time to understand the problem - Prefer multiple choice questions when possible, but open-ended is fine too - Only one question per message - Focus on understanding: expected behavior, actual behavior, reproduction steps, when it started, what changed recently **Reproducing the issue:** - Attempt to reproduce the issue before investigating code - If this is a frontend/browser issue and you see `playwright_*` tools available, use them: - `playwright_navigate` to load the page - `playwright_screenshot` to capture visual evidence - `playwright_click` / `playwright_fill` to interact with elements - `playwright_evaluate` to check console errors and state - If this is a backend/CLI issue, run the reproduction steps with Bash - If this is a test failure, run the specific failing test - If you cannot reproduce, investigate why — environment, timing, specific data, concurrency - Document the reproduction case clearly — you need it later to verify the fix **Forming hypotheses:** - Based on symptoms and code context, propose 2-3 theories about the root cause - For each theory: description, likelihood (high/medium/low), how to test it - Present theories to the user — they may have context that rules some out - Wait for user input before proceeding **Investigating:** - Work through hypotheses starting with the most likely - Read relevant code paths — use Glob and Grep to find files, Read to inspect - Check logs, trace execution, inspect state - Use Playwright for frontend state inspection if available - Present findings as you go — if Theory A doesn't pan out, explain why and move to Theory B - Once you identify the root cause with evidence, present it clearly and wait for confirmation: ``` Root cause identified: - File: [path:line] - Problem: [what's wrong] - Why: [explanation] - Evidence: [what you observed] ``` **Fixing:** - Propose the fix before implementing — the user may want a different approach - Apply the minimal change that addresses the root cause - Do not refactor unrelated code - Add a test case for the bug if it wasn't covered by tests - Add defensive checks or better error messages if appropriate **Verifying:** - Re-run the reproduction case — confirm the issue is fixed - Run relevant tests — ensure nothing else broke - Check related functionality for similar issues - If Playwright is available, take a screenshot showing the fix works ## Report After fixing, provide a clear summary: ``` Debug session complete. Issue: [brief description] Root Cause: [what was wrong] Fix: [what changed] Files modified: - [path] — [summary of changes] Verification: - [x] Issue no longer reproduces - [x] Tests pass - [x] Related functionality checked ``` ## Key Principles - **One question at a time** — do not overwhelm - **Reproduce first** — never skip this step - **Hypothesis-driven** — do not make random changes - **Evidence-based** — confirm theories with data, not guesses - **Minimal fixes** — address the root cause, not symptoms - **Verify thoroughly** — test the fix and related functionality - **Interactive** — involve the user at key decision points - **Use Playwright when available** — for frontend debugging, check your tool list for `playwright_*` tools ## When to Stop and Escalate Stop and ask the user for help if: - You cannot reproduce the issue after multiple attempts - The root cause requires domain knowledge you do not have - The fix would require architectural changes beyond a targeted patch - Multiple theories seem equally plausible and you need more context
Versions
| Repository | Score | Licence |
|---|---|---|
| This page · ratler/dream-team | 100 | MIT |
| ericrisco/rsc-harness | 100 | MIT |
| nicsuzor/academicops | 100 | MIT |
| duronext/jig | 100 | MIT |
| drosam/agents | 100 | MIT |
| rizukirr/vibekit | 100 | MIT |
| DingyangLyu/MatClaw | 100 | Apache-2.0 |
| Kanevry/session-orchestrator | 100 | MIT |
| kdlbs/kandev | 100 | AGPL-3.0 |
Database snapshot: 13.11.2026 · package version 8f3a3b57c381 (SHA-256 digest of the entire package)
Downloads in the last 30 days: 0 · ranking
Report a problem with this skill: harmful content, incorrect descriptions or an author’s removal request.
Similar skills
- 90debug-methodical
Debugging méthodique en 4 phases (reproduce → isolate → fix → verify). Use when investigating a bug, regression, flaky test, or unexpected behavior. - 90debugging
Systematic root-cause debugging: reproduce, investigate, hypothesize, fix with verification. Use when asked to "debug this", "fix this bug", "why is this failing", "troubleshoot", or mentions errors, stack traces, broken tests, flaky tests, regressions, or unexpected behavior. - 90investigate
Systematic debugging skill. Use when encountering a bug, test failure, unexpected behavior, or when asked to "investigate", "debug", "diagnose", or "figure out why". Enforces root-cause-first discipline with structured hypothesis tracking. - 90adev:debug
Context-aware systematic debugging. Checks ADRs for known issues, specs for expected behavior, and orientation for architecture context before investigating. Use when encountering any bug, test failure, or unexpected behavior. Trigger on 'fix this', 'fix the bug', 'something is… - 100workflow-fix-and-ship
One bug-fix lifecycle: triage, reproduce, debug, regression test, fix, live check, PR, optional deploy verify. Use when "fix this bug and close the ticket", "patch this and ship", or a named production error. Many reports → workflow-feedback-to-closure. - 91debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes