メニュー
Push accountability, CI monitoring after push, background agent CI verification, verification command sequencing.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
Production deployment rules, rollback-first recovery, dependency batching, CI cost awareness, and framework upgrade verification.
Always use when user asks to create, generate, draw, or design a diagram, flowchart, architecture diagram, ER diagram, sequence diagram, class diagram, network diagram, mockup, wireframe, or UI sketch, or mentions draw.io, drawio, drawoi, .drawio files, or diagram export to PNG/SVG/PDF.
Known agent error patterns -- debugging reference for tool failures, git errors, CI issues, and common mistakes. Consult when encountering unexpected behavior, tool errors, or CI failures.
Git recipes, worktree management, push sequences, branch verification, and conflict resolution patterns.
gh CLI patterns, JSON field discovery, PR check interpretation, label management, merge settings verification.
macOS-specific patterns: launchd agent configuration, brew vs pip, zsh regex quirks, file descriptor limits.
SOC 職業分類に基づく
| name | CI Workflow |
| description | Push accountability, CI monitoring after push, background agent CI verification, verification command sequencing. |
Use the branch that is currently under CI verification. In a
develop/main topology this is often develop. In a main-only repo
it is usually the temporary branch or PR branch being validated before
merge.
Wrong -- push and move on:
git push origin <branch-under-test>
# Start next task immediately, never check CI
Right -- spawn background agent to monitor CI:
git push origin <branch-under-test>
# Background agent:
gh run list --branch <branch-under-test> --limit 1
# If CI fails: investigate with gh run view <id> --log-failed
# Fix and re-push. The push isn't done until CI is green.
Wrong -- run typecheck, lint, test as parallel tool calls:
# Parallel call 1: pnpm run typecheck
# Parallel call 2: pnpm run lint
# Parallel call 3: pnpm run test
# If one fails, all parallel calls are killed (Error #1)
Right -- chain sequentially with semicolons:
pnpm run typecheck 2>&1; pnpm run lint 2>&1; pnpm run test 2>&1
Wrong -- commit first, discover failures from pre-commit hook:
git commit -m "feat: add feature"
# Pre-commit hook fails: lint errors, type errors
Right -- run checks before committing:
pnpm run typecheck 2>&1; pnpm run lint 2>&1
git add <files> && git commit -m "feat: add feature"
Wrong -- change tsconfig and continue coding:
# Edit tsconfig.json
# Continue implementing next feature
# Discover 200 type errors at commit time
Right -- run full test suite immediately after config changes:
# Edit tsconfig.json
pnpm run typecheck 2>&1; pnpm run lint 2>&1; pnpm run test 2>&1
# Fix any breakage before proceeding