{"enrichment":{"faq":[{"a":"Investigate teaches a five-phase debugging process: collect evidence first, form ranked hypotheses from that evidence, verify each hypothesis methodically, apply a minimal fix only after confirming root cause, then verify the fix resolves the issue. This structured approach prevents guess-and-fix cycles by enforcing evidence gathering before hypothesis formation and stopping endless speculation with a three-strike rule that forces re-collection of evidence.","q":"How do I debug a bug systematically with Investigate?"},{"a":"Investigate's root cause analysis follows a disciplined sequence: gather concrete evidence about the failure, rank multiple hypotheses by likelihood based on that evidence, test each hypothesis systematically rather than guessing, confirm which one is actually the root cause, and only then apply a minimal targeted fix. This methodology ensures you diagnose code issues properly instead of applying premature fixes to symptoms.","q":"What is the root cause analysis workflow in Investigate?"},{"a":"Investigate guides you to investigate unexpected behavior by collecting evidence first, then forming ranked hypotheses about what caused it. You verify each hypothesis methodically before moving to the next, avoiding the pitfall of skipping phases. The three-strike rule prevents endless speculation\u2014after three failed hypotheses, you must re-collect evidence rather than continue guessing, ensuring you develop debugging discipline.","q":"How does Investigate help me investigate unexpected behavior?"},{"a":"Yes. Investigate's structured debugging methodology eliminates guessing by enforcing an evidence-first approach. You collect concrete data about the failure, form ranked hypotheses from that evidence, then verify each one systematically. Only after confirming the actual root cause do you apply a fix. This prevents the common pitfall of skipping phases and teaches debugging discipline that avoids premature fixes and wasted troubleshooting.","q":"Can Investigate help me debug without guessing?"},{"a":"Investigate enforces a five-phase process that separates evidence gathering from hypothesis formation, preventing guess-and-fix cycles. Its three-strike rule stops endless speculation by forcing re-collection of evidence after three failed hypotheses. By requiring verification before applying any fix, Investigate ensures you diagnose root cause properly and apply only minimal, targeted solutions\u2014teaching debugging best practices that avoid common pitfalls.","q":"What makes Investigate's debugging approach different?"},{"a":"Investigate's final phase verifies resolution after you apply the minimal fix. This completes the five-phase workflow: evidence collection, hypothesis ranking, verification of root cause, minimal fix application, and confirmation that the fix actually resolves the original failure. This systematic approach ensures your debugging process is complete and the issue is truly resolved, not just masked.","q":"How does Investigate verify a fix resolves the issue?"}],"shadow_tags":["root-cause-analysis","hypothesis-driven-debugging","error-diagnosis","systematic-troubleshooting","regression-testing","evidence-collection","code-investigation","test-failure-resolution","bug-reproduction"],"summary_rewrite":"Investigate guides you through a disciplined five-phase debugging process: collect evidence, form ranked hypotheses, verify each one methodically, apply a minimal fix only after confirming root cause, then verify resolution. The workflow prevents guess-and-fix cycles by enforcing evidence gathering before hypothesis formation and a three-strike rule that forces re-collection rather than endless speculation."},"files":[{"bytes":2258,"path":"skills/skills/investigate/SKILL.md","sha256":"5f1dbb250e4570d381667b79706f5e28a445669c86427024531448175fe29042","url":"https://skillfed.io/files/blueberrycongee/termcanvas/investigate/7d84702b/SKILL.md"}],"id":"blueberrycongee/termcanvas/investigate","links":{"html":"https://skillfed.io/blueberrycongee/termcanvas/investigate","md":"https://skillfed.io/blueberrycongee/termcanvas/investigate.md","repo":"https://github.com/blueberrycongee/termcanvas"},"meta":{"agents_supported":[],"first_seen":"2026-07-28","forks":29,"language":"TypeScript","last_updated":"2026-05-31","license":"MIT","name":"investigate","publisher":"blueberrycongee","stars":374},"relations":{"categories":["debugging","quality-assurance","software-engineering"],"similar":[{"id":"FlorianBruniaux/claude-code-ultimate-guide/investigate"},{"id":"blueberrycongee/termcanvas/using-termcanvas"},{"id":"catlog22/Claude-Code-Workflow/investigate"},{"id":"blueberrycongee/termcanvas/qa"},{"id":"blueberrycongee/termcanvas/challenge"},{"id":"blueberrycongee/termcanvas/code-review"},{"id":"blueberrycongee/termcanvas/security-audit"},{"id":"garrytan/gstack/gstack-openclaw-investigate"},{"id":"blueberrycongee/termcanvas/hydra"},{"id":"softspark/ai-toolkit/debug"}]},"slug":{"owner":"blueberrycongee","repo":"termcanvas","skill":"investigate"},"version":"7d84702b"}