본문으로 건너뛰기
cccskills
무료GitHub에 공개

fix

Diagnose a Kandev bug, reconcile its requirements and system design, create a reviewable fix plan and work orders, then hand off for explicit TDD implementation.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev스타 1,024(GitHub 1,024 + cccskills 0)

실행 점수 미검증

설치 방법 보기

포함된 파일(1)

  • SKILL.md5.8 KB

SKILL.md(원문)

설치하기 전에 에이전트에게 전달되는 지시 내용을 확인할 수 있습니다.

Fix

Use the same durable workflow as feature work. Diagnose the defect first. Then reconcile specifications and create the fix plan and work orders before you change production code.

Core flow

Evidence -> Root cause -> Requirement conformance -> System-design check -> Fix plan + work orders -> Handoff -> Explicit implementation request -> TDD -> PR review

Do not patch production code before the planning checkpoint unless the user explicitly opts out.

Phase 0: Evidence and root cause

When a bug comes from an issue tracker, read the canonical issue and each image attachment. Reproduce the behavior with existing tests, a read-only trace, or a minimal temporary reproduction.

State:

  • The incorrect behavior and its conditions.
  • The root cause.
  • The smallest reliable reproduction.
  • The intended regression-test level and path.

If the cause remains uncertain, stop and ask the user.

Phase 1: Reconcile specifications

Run the /interview-me assumption check for intended behavior and repair scope. Reuse active acceptance criteria for clear regressions. Ask only about material choices that evidence and prior instructions do not settle.

Read docs/specs/README.md, the owning system index, and the relevant requirement and system-design documents. Use the legacy catalog only when the system has not migrated. Read the owning system README.md and use this command for discovery:

python3 scripts/list-docs.py specs --format paths

Add --kind legacy when the system has not migrated.

Search adjacent systems before you create or move an artifact. A UI symptom does not make the repair UI-owned. Update the system that owns the failed contract, and include observable UI recovery there when applicable.

Classify the bug:

  1. Implementation violates an active acceptance criterion. Reference the existing AC-* ID. Do not create or rewrite a requirement.
  2. The intended behavior is missing from requirements. Add the smallest requirement and acceptance criteria that define it.
  3. The intended product behavior changes. Amend or supersede the affected requirement. Record the compatibility effect.
  4. The technical design is wrong or incomplete. Update the system design. Keep observable behavior in requirements.

Do not create a standalone repair specification. The requirement is the durable behavioral source. The plan and work orders record this repair.

For predicates over collections or inventory, state the quantifier explicitly: distinguish existential conditions such as “at least one live row” from universal conditions such as “all rows.” When behavior depends on that distinction, require a mixed-state regression containing both live and failed or deleted rows.

Do not create parallel feature and UI requirements for one repair. A separate UI requirement is valid only when the repair changes an independent reusable UI contract.

Use /record when the correction establishes a durable boundary, ownership rule, public contract, persistence rule, or security invariant with meaningful alternatives.

Phase 2: Create the fix package

Create docs/plans/<fix-slug>/plan.md and sibling work orders through /plan.

The package must:

  • Reference the affected REQ-* and AC-* IDs.
  • Reference the applicable system design.
  • State the confirmed root cause.
  • Name the regression test that fails before the correction.
  • Name exact files, dependencies, acceptance conditions, and commands.
  • Use dependency waves only when they clarify implementation order.
  • For rendered UI changes, include ASCII previews in the plan and relevant UI work orders per docs/specs/guide/plans-and-work-orders.md#ascii-ui-previews. Show before/after when it explains the defect and correction.

Keep work orders sequential by default. A wave does not authorize delegation.

Phase 3: Design-package handoff

Before you change production or permanent test code, report:

  • Root cause and reproduction evidence.
  • Requirement IDs and system-design paths.
  • Plan and work-order paths.
  • Dependency order and exact commands.
  • Risks and exclusions.
  • For UI changes, a compact inline ASCII preview of the correction, including phone composition when different, and links to the full plan/work-order views.

Then end the turn. Do not ask the user to approve the package or switch models. Wait for a later explicit implementation request.

Phase 4: Implement with TDD

After the explicit request:

  1. Change the work order to in_progress.
  2. Write and run the regression test.
  3. Make sure that the test fails for the expected reason.
  4. Implement the minimum correction.
  5. Run the work order's exact commands.
  6. Change the work order to done and synchronize plan.md.

For persisted defaults or coupled settings, inspect every write and reset path. Cover explicit empty and null values when they affect the contract.

If the user authorizes subagents, obey /planner-orchestration. Use native delegation, compact work-order context, no recursive delegation, and no full-history fork.

Phase 5: PR review

After all work-order checks pass, commit, push, and open the PR. Do not add an automatic local simplify, QA, broad review, security review, or verification pass.

Use /pr-fixup only for a CI error or an actionable reviewer finding. Run the affected work-order check after a correction.

Stop conditions

Stop when the root cause is uncertain, specifications and code disagree, the repair needs a new material design, or the same targeted check fails three times.

Final report

Report the root cause, requirement IDs, system-design path, plan and work-order paths, changed files, commands, results, authorized subagents, and PR status.

리뷰

아직 리뷰가 없습니다. 사용해 본 소감을 남겨 주세요.

댓글

로그인 없이 쓸 수 있습니다. 사용 후기, 질문, 작동하지 않은 환경 등을 편하게 남겨 주세요.

아직 댓글이 없습니다. 첫 댓글을 남겨 보세요.

보내면 바로 공개됩니다. 이메일 주소나 전화번호 같은 개인 정보는 쓰지 마세요. URL은 2개까지입니다.

0 / 1,000자

이 양식은 reCAPTCHA로 보호되며 Google 개인정보처리방침 및 서비스 약관이 적용됩니다.

같은 저장소의 스킬

acp-debug

무료

Debug an ACP agent CLI by spawning it, speaking raw JSON-RPC, and capturing every frame to a JSONL file. Use when the user asks to probe an agent's capabilities, compare agents, test a prompt against an agent, inspect raw ACP wire frames, or investigate why an agent fails to initialize.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

Add a new third-party integration (Jira/Linear-style) — per-workspace credentials, 90s auth-health poller, settings page, link/import buttons. Use when scaffolding a new external service integration.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

Delete local branches whose remote has been deleted ([gone]), including their worktrees. Use for branch cleanup after merging PRs.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

Review changed code for quality, security, and architecture compliance. Use only when the user explicitly requests local review or a PR finding requires it.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

commit

무료

Stage and commit changes using Conventional Commits. Use when there are dirty/staged files to commit, the user says "commit", or before pushing a PR.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

Curate the right project context before coding or debugging. Use when starting a new session, switching areas of the codebase, output quality is drifting, a task spans backend/frontend/docs, or external instructions need to be reconciled with Kandev conventions.

한국어 요약은 준비 중입니다. 원문 설명을 표시하고 있습니다.

kdlbs/kandev1천2026년 10월 26일 업데이트

kdlbs의 스킬 모두 보기

이 스킬의 문제 신고