hunt
The hunt skill provides a structured framework for diagnosing and resolving software defects, including errors, crashes, regressions, and test failures. It enforces a root-cause-first methodology, requiring users to identify the specific source of a problem before applying any code changes. The skill includes specialized modes for bisecting regressions, handling visual or screenshot-reported defects, and performing scope analysis to prevent similar bugs across a codebase.
Is hunt safe to install?
Review the source first: our audit of hunt's source files found 15 shell commands, 0 external URLs, file reads and writes (high risk). Every command and URL listed appears verbatim in the skill's source. The skill executes shell commands, including git operations and system diagnostic tools, and performs file system operations such as grepping for patterns and reading project files.
How we audit skills: our security review methodology.
Who is this skill for?
Developers and engineers who need to debug complex software issues, investigate regressions, or resolve persistent defects in a systematic and evidence-based manner.
What can you do with it?
- Diagnosing root causes for application crashes or runtime errors.
- Investigating regressions where previously functional code fails after an update.
- Analyzing visual or rendering defects reported via screenshots.
- Performing git bisect operations to identify the commit that introduced a bug.
- Conducting scope analysis to find and fix similar bug patterns across a repository.
- Troubleshooting performance issues or native application freezes.
How good is this skill?
Quality score: 5/10. The skill provides a highly disciplined and rigorous framework for debugging. It emphasizes evidence-based reasoning and provides clear, actionable protocols for various failure scenarios.
What does the skill file contain?
# Hunt: Diagnose Before You Fix Prefix your first line with 🥷 inline, not as its own paragraph. **Update check (non-blocking).** Once per conversation, run `bash <skill-base-dir>/scripts/check-update.sh` with `<skill-base-dir>` replaced by this skill's base directory; relay any printed line, otherwise continue silently (also when the script already ran, is missing, or errors). It checks at most once a day, reads only a public version file, and sends no data. A patch applied to a symptom creates a new bug somewhere else. ## Outcome Contract - Outcome: the root cause is identified before a...
Frequently asked questions
When should I use the hunt skill?
Use it when you encounter errors, crashes, regressions, failing tests, broken behavior, or screenshot-reported defects. It is not intended for code reviews or implementing new features.
What is the requirement before applying a fix?
You must identify the root cause and state it in one sentence, naming a specific file, function, line, or condition. You must also provide evidence such as a source trace, logs, or a reproducible test case.
How does the skill handle regressions?
It provides a Bisect Mode to identify the culprit commit. It requires protecting the worktree, defining a non-interactive pass/fail test, and using git bisect to isolate the regression.
What happens if I cannot find the root cause after multiple attempts?
The skill mandates a hard stop after three failed hypotheses. You must use the Handoff format to document what was checked, what was ruled out, and what remains unknown.
Related skills
systematic-debugging
173.2KSoftware engineers and AI agents tasked with resolving bugs, test failures, or unexpected system behaviors
A structured methodology for debugging technical issues by prioritizing root cause identification over symptom-based patching. It mandates a four-phase process: root cause investigation, pattern analysis, hypothesis testing, and implementation, while providing specific protocols for handling persistent failures and architectural concerns.
convex-performance-audit
79.2KDevelopers working with Convex applications who need to optimize slow features, resolve high-resource usage, or address transaction contention
The convex-performance-audit skill provides a structured workflow for diagnosing and resolving performance issues in Convex applications, including read amplification, OCC conflicts, subscription costs, and function limits.
momentic-result-classification
29.5KDevelopers and QA engineers using the Momentic testing framework to triage and debug automated test failures
The momentic-result-classification skill provides tools to analyze, categorize, and explain test failures within the Momentic end-to-end testing framework. It enables users to investigate test run metadata, step-level results, and historical run data to identify root causes of failures.
flutter-fix-layout-issues
21.1KFlutter developers encountering layout exceptions during UI development
This skill provides a diagnostic and resolution workflow for common Flutter layout errors, including RenderFlex overflows and unbounded constraint violations.