Skills DirectorySkills Directory
SkillsLearnSecurityCategoriesDocsCommunityBlog
Sign InSubmit Skill
Skills Directory

Security-tested agent skills for Claude, coding agents, and AI workflows.

Directory

  • Browse Skills
  • All Skills A–Z
  • Claude Skills
  • Claude Code Skills
  • Agent Skills
  • Categories
  • Submit a Skill

Learn

  • Learn Hub
  • Install Claude Skills
  • Write SKILL.md
  • Skills vs MCP
  • Directories Compared

Security

  • Security
  • Methodology
  • Secure Claude Skills
  • Security Badges

Company

  • About
  • Community
  • Blog
  • API Docs
  • Advertise

2026 Skills Directory. All rights reserved.

Back to skills

Fix Bug

ASecurity

Maintenance shortcut — pick highest priority bug, find root cause, fix, verify

8 stars
0 votes
0 copies
0 views
Added 9/20/2026
testinggitapi

Works with

api

Security Analysis

A100/100

Scanned 9/20/2026

Install to Claude Code

$npx -y skills add tstapler/dotfiles --skill fix-bug --agent claude-code

Installs into .claude/skills of the current project.

Are you the author of Fix Bug?

Add the live security badge to your README — it updates automatically with every re-scan.

Security grade badge for Fix Bug
[![Security: A — Skills Directory](https://www.skillsdirectory.com/api/skills/tstapler-fix-bug-dotfiles/badge)](https://www.skillsdirectory.com/skills/tstapler-fix-bug-dotfiles)

More formats (shields.io, HTML) on the badges page.

Download Zip
Files
SKILL.md
---
description: Maintenance shortcut — pick highest priority bug, find root cause, fix, verify
user-invocable: true
---

# sdd:fix-bug

Three-phase maintenance workflow: root cause → fix → verify.

## Instructions

1. **Follow [SETUP.md](../skills/SETUP.md)** — verify prerequisites.

2. **Select the bug.** If an argument is provided, use it. Otherwise scan `docs/bugs/open/` and select the highest priority open bug.

3. **Phase A — Root cause.**

   Investigate the root cause before proposing a fix — a fix aimed only at the symptom will resurface.

   Before proposing any fix:
   - Read the full error, stack trace, and surrounding context
   - Reproduce the failure locally or identify the exact condition that triggers it
   - Check recent commits for related changes (`git log --oneline -20`)
   - Form a hypothesis: "The root cause is X because Y"

4. **Phase B — Fix.**

   - Fix the root cause, not the symptom
   - Write a regression test that would have caught this bug

5. **Phase C — Verify.**

   Claim the bug is fixed only after running the test and showing the passing output.

   Run the relevant test(s) using the appropriate test command for the stack.
   Show the full output.

6. **Phase D — Reflect (fix the class, not the instance).**

   A regression test proves *this* bug won't come back. It doesn't prove the *shape* of bug
   won't recur elsewhere. Before closing out, apply the `quality:reflect-and-fix` taxonomy to
   the root cause you found in Phase A:

   - Classify it: Semantic/Intent, Framework Pattern Misuse, API Contract Gap, Type Safety
     Gap, Integration Gap, or Dependency/Build Gap.
   - Ask: what's the *earliest* point on the enforcement ladder (compile-time type →
     lint/static → unit test → integration test → checklist) that would have caught this?
     If the Phase B regression test is already the earliest achievable level, say so and move
     on — don't manufacture enforcement for its own sake.
   - If a type change, lint rule, or ast-grep/semgrep pattern could have caught it earlier
     than a runtime test, implement that too (invoke `quality:reflect-and-fix` for the full
     4-phase version if the bug is non-trivial; apply the ladder inline if it's a quick call).
   - **Recurring-shape check**: is this the Nth bug with the same underlying pattern — e.g.
     "a spawn call silently no-ops instead of erroring," "a sweep meant to catch dead state
     excludes the exact case it should catch," "an event is lost across a restart with no
     catch-up path"? If so, this is a systemic gap, not an isolated bug — the fix must close
     the whole class (e.g. a shared helper with the invariant enforced once, a lint rule
     banning the unsafe pattern repo-wide, a structural test asserting the sweep's exclusion
     guard can't self-defeat) rather than patching this one call site. Say explicitly in the
     bug doc which recurring shape this is, so a future bug-fix or audit doesn't re-derive it
     as new.

7. **Update the bug document** in `docs/bugs/open/` → move to `docs/bugs/resolved/`, including
   the Phase D classification and any recurring-shape note.

8. Output:
   ```
   ✅ Bug fixed and verified

   Root cause: <one sentence>
   Fix: <one sentence>
   Regression test: <test name>
   Systemic fix: <enforcement added beyond the regression test, or "none needed — test is the
     earliest achievable level">
   Recurring shape: <name of the pattern if this is the Nth instance, or "none identified">
   ```

Attribution

tstaplertstapler
View sourceMore from tstapler →
SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Is this your skill, or is something wrong with this listing? Request removal or report an issue. Author removals are honored within 72 hours.

Comments (0)

No comments yet. Be the first to comment!

SSkills DirectorySkills Directory

Your tool, in front of Claude Code builders.

3 founder slots · $299/mo · GSC-verified traffic · sponsors can never buy grades.

See placements

Related Skills

Screen Reader Testing

Practical guide to testing web applications with screen readers for comprehensive accessibility validation.

393431 votes

Python Testing

使用pytest、TDD方法、夹具、模拟、参数化和覆盖率要求的Python测试策略。

2456590 votes

Tdd Workflow

在编写新功能、修复错误或重构代码时使用此技能。强制执行测试驱动开发,包含单元测试、集成测试和端到端测试,覆盖率超过80%。

2456590 votes

Springboot Tdd

使用JUnit 5、Mockito、MockMvc、Testcontainers和JaCoCo进行Spring Boot的测试驱动开发。适用于添加功能、修复错误或重构时。

2456590 votes

Eval Harness

克劳德代码会话的正式评估框架,实施评估驱动开发(EDD)原则

2456590 votes
View all in testing →