원클릭으로
rai-story-implement
Execute plan tasks with TDD and validation gates. Use after story plan.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Execute plan tasks with TDD and validation gates. Use after story plan.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Interactive adapter setup for Jira and Confluence. Detects available backends, discovers projects/spaces, generates validated YAML config. 3-4 questions max.
Evaluate design proportionality using Beck's four rules. Use after implementation.
Root cause analysis using the method best suited to the bug. Phase 3 of bugfix pipeline.
Push branch, create MR, verify artifacts complete. Phase 7 of bugfix pipeline.
Execute fix tasks with TDD and all validation gates. Phase 5 of bugfix pipeline.
Decompose fix into atomic TDD tasks. Phase 4 of bugfix pipeline.
| name | rai-story-implement |
| description | Execute plan tasks with TDD and validation gates. Use after story plan. |
Execute the implementation plan task by task with TDD, producing verified code that passes all gates.
When to use: After /rai-story-plan has produced a plan document.
Prerequisite: Plan must exist at work/epics/e{N}-{name}/stories/{story_id}/plan.md. Run /rai-story-plan first if missing.
Inputs: Implementation plan, project guardrails (from graph context).
Before starting Step 1, you MUST execute the PRIME protocol:
.raise/rai/learnings/rai-story-plan/{work_id}/record.yaml.rai graph query. If graph is unavailable, note and continue.JIT: Before loading context, query graph for implementation patterns in affected modules →
aspects/introspection.md § JIT Protocol
Load the implementation plan and query relevant patterns:
rai graph query "testing coverage type annotations" --types pattern,guardrail --limit 5
If a design document exists, restate the design intent in 2-3 sentences and confirm with the human before proceeding. One unvalidated assumption can waste an entire task cycle.
For the next uncompleted task in plan order:
Follow project rules, guardrails, and established patterns.
Run the verification defined in the plan. Resolve commands using this priority chain:
.raise/manifest.yaml for project.test_command, project.lint_command, project.type_check_command — if set, use directlyproject.project_type in manifest, or scan file extensions| Language | Test | Lint | Format | Type Check |
|---|---|---|---|---|
| Python | uv run pytest --tb=short | uv run ruff check src/ tests/ | uv run ruff format --check src/ tests/ | uv run pyright |
| TypeScript | npx vitest run | npx eslint src/ | npx prettier --check src/ | npx tsc --noEmit |
| JavaScript | npx vitest run | npx eslint src/ | npx prettier --check src/ | — |
| C# | dotnet test | dotnet format --check | — | dotnet build |
| Go | go test ./... | golangci-lint run | gofmt -l . | go vet ./... |
| PHP | vendor/bin/phpunit | php-cs-fixer check | — | vendor/bin/phpstan |
| Dart | flutter test | dart fix --dry-run | dart format --set-exit-if-changed . | dart analyze |
The manifest always wins when present. The table is a fallback.
Run ALL four gates (test + lint + format + type check) after each task, not just the one mentioned in the plan. The goal is to catch errors locally before they reach CI.
If verification fails: fix and re-verify (max 3 attempts before escalating).
work/epics/.../stories/{story_id}/progress.md)| Condition | Action |
|---|---|
| More tasks remain | Return to Step 2 |
| All tasks complete | Run full gate check, present summary |
| Task blocked | Document blocker, escalate to human |
| Item | Destination |
|---|---|
| Implemented code | Per project architecture |
| Progress log | work/epics/.../stories/{story_id}/progress.md |
| Next | /rai-story-review |
gates/gate-code.md/rai-story-reviewreferences/progress-template.md