investigate
A systematic debugging process that prioritizes evidence-based root-cause analysis over speculative fixes.
Install
mkdir -p .claude/skills/investigate-yokoyamaryota && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/15003" && unzip -o skill.zip -d .claude/skills/investigate-yokoyamaryota && rm skill.zipInstalls to .claude/skills/investigate-yokoyamaryota
Activation
This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.
Root-cause-first debugging workflow. Use for bugs, regressions, stack traces, production issues, or any case where the real failure path must be understood before edits are proposed.Key capabilities
- →Reproduce the issue and capture exact symptoms
- →Trace the real execution path
- →Form the smallest plausible root-cause hypothesis
- →Verify the hypothesis against code or runtime evidence
- →Propose or implement a fix
How it works
This skill defines a sequential workflow that starts with reproducing an issue and ends with verifying a hypothesis before proposing a fix.
Inputs & outputs
When to use investigate
- →Debugging production issues
- →Investigating regressions
- →Analyzing stack traces
- →Root cause analysis
- →Verifying bug fixes
About this skill
Investigate
Rules
- no speculative fixes
- gather evidence before changing code
- separate observations, hypotheses, and fixes
Workflow
- Reproduce the issue and capture exact symptoms.
- Trace the real execution path.
- Form the smallest plausible root-cause hypothesis.
- Verify the hypothesis against code or runtime evidence.
- Only then propose or implement a fix.
When not to use it
- →When a speculative fix is desired
- →When code changes are preferred before gathering evidence
- →When observations, hypotheses, and fixes are not separated
Limitations
- →This skill does not permit speculative fixes
- →This skill requires evidence gathering before code changes
- →This skill separates observations, hypotheses, and fixes
How it compares
This workflow enforces a structured, evidence-based approach to debugging, unlike directly attempting fixes.
Compared to similar skills
investigate side by side with the closest alternatives in the catalog.
| Skill | Installs | Updated | Safety | Difficulty |
|---|---|---|---|---|
| investigate (this skill) | 0 | 4mo | No flags | Intermediate |
| corpus-management | 0 | 2mo | Review | Advanced |
| python-testing-patterns | 77 | 2mo | Review | Intermediate |
| chrome-devtools | 41 | 7mo | Review | Intermediate |
Try saying
Example prompts that trigger this skill in your AI assistant.
You might also like
corpus-management
xsscx
>
python-testing-patterns
wshobson
Implement comprehensive testing strategies with pytest, fixtures, mocking, and test-driven development. Use when writing Python tests, setting up test suites, or implementing testing best practices.
chrome-devtools
mrgoonie
Browser automation, debugging, and performance analysis using Puppeteer CLI scripts. Use for automating browsers, taking screenshots, analyzing performance, monitoring network traffic, web scraping, form automation, and JavaScript debugging.
bats
OleksandrKucherenko
Bash Automated Testing System (BATS) for TDD-style testing of shell scripts. Use when: (1) Writing unit or integration tests for Bash scripts, (2) Testing CLI tools or shell functions, (3) Setting up test infrastructure with setup/teardown hooks, (4) Mocking external commands (curl, git, docker), (5) Generating JUnit reports for CI/CD, (6) Debugging test failures or flaky tests, (7) Implementing test-driven development for shell scripts.
browser-daemon
noiv
Persistent browser automation via Playwright daemon. Keep a browser window open and send it commands (navigate, execute JS, inspect console). Perfect for interactive debugging, development, and testing web applications. Use when you need to interact with a browser repeatedly without opening/closing it.
performance-profiling
davila7
Performance profiling principles. Measurement, analysis, and optimization techniques.