systematic-debugging
Skill16 starsChanged 4 days agoMITView on GitHub
Phase-based root cause investigation: reproduce, observe, hypothesize, bisect, verify. TRIGGER when: investigating a bug report, diagnosing unexpected behavior, or doing root cause analysis. SKIP: looking up a known error first (use sofa-search); writing the regression test (use testing-strategy); live-repro watcher (use watch-patterns).
190 other projects use a version of this.
Adopted by
Most recently taken up by
- DemonDamon/AgenticXtheir own version, first seen Nov 9, 2026
- chrysogonus/power-agentstheir own version, first seen Nov 8, 2026
- willy92wins/DayZ-Modding-Knowledge-Packtheir own version, first seen Nov 8, 2026
- roundpilot/superpowers-antigravitytheir own version, first seen Nov 8, 2026
- REPOZY/superpowers-optimizedtheir own version, first seen Nov 8, 2026
- lingyired/status-triotheir own version, first seen Nov 8, 2026
Used with
Projects with it also use chinese-code-review, chinese-commit-conventions and 10 more.
Other files people use alongside this one.
Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
Leodorareluctant259Skillchinese-code-review
中文代码审查规范——在保持专业严谨的同时，用符合国内团队文化的方式给出有效反馈
Also in 8 projects that use this
- Leodorareluctant259Skill
chinese-commit-conventions
中文 Git 提交规范 — 适配国内团队的 commit message 规范和 changelog 自动化
Also in 8 projects that use this
obraSkilldiagnosing-superpowers
Use when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.
Also in 20 projects that use this
mjunaidcaSkillbuilding-chatgpt-apps
Guides creation of ChatGPT Apps with interactive widgets using OpenAI Apps SDK and MCP servers. Use when building ChatGPT custom apps with visual UI components, embedded widgets, or rich interactive experiences. Covers widget architecture, MCP server setup with FastMCP, response metadata, and Developer Mode configuration. NOT when building standard MCP servers without widgets (use building-mcp-servers skill instead).
Also in 4 projects that use this
- mjunaidcaSkill
building-chat-interfaces
Build AI chat interfaces with custom backends, authentication, and context injection. Use when integrating chat UI with AI agents, adding auth to chat, injecting user/page context, or implementing httpOnly cookie proxies. Covers ChatKitServer, useChatKit, and MCP auth patterns. NOT when building simple chatbots without persistence or custom agent integration.
Also in 4 projects that use this
- +2 more from Leodorareluctant259
- +5 more from mjunaidca
Other versions190 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 190
- obra/superpowers★ 314ksame name
- NousResearch/hermes-agent★ 262ksame name
- open-metadata/OpenMetadata★ 16ksame name
- jnMetaCode/superpowers-zh★ 8.4ksame name
- vudovn/ag-kit★ 8.2ksame name
- vudovn/antigravity-kit★ 8.2ksame name
- openai/plugins★ 7.8ksame name
- ChrisWiles/claude-code-showcase★ 6.1ksame name
- farm-fe/farm★ 5.6ksame name
- microsoft/FluidFramework★ 4.9ksame name
180 more not listed
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository39 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- agent-commsSkill
- agent-memorySkill
- api-designSkill
- context-engineeringSkill
- database-optimizationSkill
- distillSkill
- docker-templatesSkill
- error-handlingSkill
- frontend-designSkill
- github-actions-templateSkill
- git-operationsSkill
- logging-standardsSkill
- mcp-toolsSkill
- monitoring-observabilitySkill
- mui-stylingSkill
- necessity-firstSkill
- pattern-recognitionSkill
- planning-methodologySkill
- python-patternsSkill
- quality-validationSkill
- react-patternsSkill
- research-methodologySkill
- reverse-engineer-anythingSkill
- security-review-checklistsSkill
- semantic-memory-mcpSkill
- skill-authoringSkill
- sofa-searchSkill
- spec-designSkill
- spec-developSkill
- spec-researchSkill
- spec-reviewSkill
- spec-workflowSkill
- state-managementSkill
- testing-strategySkill
- ui-ux-pro-maxSkill
- verification-before-completionSkill
- watch-patternsSkill
- worktree-managementSkill
The file
We split this skill into the rules it gives an agent
1 rule. 0 are common, 1 is distinctive, and 0 are very specific to this project.
0
Common
Most of the 177 edited versions we read keep these.
1
Distinctive
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 25
Do NOT apply this skill for:
“Common” is measured: kept by at least 6 in 10 of the 177 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Systematic Debugging Skill
Purpose
A repeatable root cause investigation procedure. A bug is not understood until it can be reproduced on demand and the causal chain is traced back to a fixable defect. This skill encodes the phase order and the discipline (one variable at a time, bisect instead of scan, verify by reverting) that keeps an investigation from turning into guesswork.
When to Apply
Apply this skill when:
- Investigating a reported bug or unexpected behavior
- Diagnosing a failing test, crash, or incorrect output
- Performing root cause analysis before recommending a fix
- An error message, stack trace, or log output needs to be traced to its origin
Do NOT apply this skill for:
- Writing the regression test that prevents recurrence — use
testing-strategy - Constructing the poll/watch loop used to observe a live reproduction — use
watch-patterns - Looking up a known/unfamiliar error before investigating from scratch — use
sofa-search
Investigation Phases
Work the phases in order. Do not skip ahead to a fix before phase 6.
| Phase | Action | Exit Criterion |
| 1. Reproduce first | Establish a minimal, reliable repro before forming any hypothesis | Bug reproduces on demand, deterministically |
| 2. Observe before theorizing | Read the actual error text, stack trace, and logs verbatim | Symptom described using only observed evidence, zero assumptions |
| 3. Hypothesize | State one falsifiable hypothesis and the prediction that would confirm or refute it | Hypothesis has a concrete, testable prediction |
| 4. Test one variable | Change exactly one thing, run the test, compare to the prediction | Prediction confirmed or refuted; refuted hypotheses are discarded, not recycled |
| 5. Bisect the search space | Narrow via git bisect, binary search on input, or disabling half the config — not linear code reading | Candidate range halves each round |
| 6. Trace backward | Walk the call stack from the crash site back to the original trigger | Triggering call/input identified, not just the frame that crashed |
| 7. Root cause, not proximate cause | Keep asking "why" until the chain reaches something you can actually fix | Chain terminates at a fixable defect (bad state, wrong assumption, missing check) |
| 8. Verify by reverting | Revert the fix and confirm the symptom returns; reapply and confirm it clears | Causation proven in both directions |
Bisection Techniques by Search Space
| Search space | Technique |
| Regression across commits | git bisect between known-good and known-bad revisions |
| Large/complex input triggers failure | Binary search: halve the input, check which half still fails |
| Behavior depends on config/flags | Disable half the flags/config, narrow to the culprit |
| Unclear which layer owns the bug | Bisect the call stack: instrument the midpoint, check state there |
Hypothesis Discipline
| Rule | Why |
| One hypothesis at a time | Multiple simultaneous changes make it impossible to attribute the outcome |
| Prediction must be falsifiable | An untestable hypothesis cannot be confirmed or ruled out |
| Never pattern-match a remembered bug onto new symptoms | Confirmation bias skips the actual evidence in phase 2 |
| A refuted hypothesis is discarded, not adjusted and reused | Adjusting mid-test invalidates the prior result |
Anti-Patterns
| Anti-Pattern | Problem | Instead |
| Forming a hypothesis before reproducing | Guessing wastes cycles on an unconfirmed bug | Reproduce deterministically first |
| Pattern-matching to a remembered similar bug | Misses the actual cause, confirmation bias | Read the actual error/stack/logs fresh (phase 2) |
| Changing multiple things per test | Cannot attribute the result to any one change | Change exactly one variable per test |
| Reading code top-to-bottom to find the defect | Slow, linear, scales poorly | Bisect the search space (phase 5) |
| Stopping investigation at the crash site | Fixes the symptom, not the defect | Trace backward to the original trigger (phase 6) |
| Declaring root cause without reverting the fix | Causation asserted, not proven | Revert, confirm symptom returns, reapply (phase 8) |
Quality Checklist
- Bug reproduces reliably, and reproduction happened before any hypothesis was formed
- Investigation cites actual error text/stack/logs, not assumptions or memory of a similar bug
- Each hypothesis had a falsifiable prediction, and only one variable changed per test
- The search space was bisected, not scanned linearly
- The causal chain was traced backward from the symptom through the call stack to the trigger
- The identified cause is a fixable root cause, not a proximate symptom
- The fix was verified by reverting it and confirming the symptom returns
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.