Menu
Execute materialized fullstack task files with layer-aware agent routing
The review-first prompt is selected by default. You can switch to a direct command or download a local copy.
Read SKILL.md and any companion files shown by SkillsMP before deciding whether to install.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
A direct command skips the review prompt. Inspect the source before running it.
npx skills add https://github.com/shinpr/claude-code-workflows --skill recipe-fullstack-buildThe command stays on one line. Scroll horizontally to inspect it before copying.
Prefer a local copy? Download the files currently available to SkillsMP.
Determines which of PRD, ADR, UI Spec, Design Doc, and Work Plan a change requires, and where each is stored. Use when deciding documentation scope, or when creating or reviewing a technical document.
Execute from repository evidence through applicable UI Spec and optional ADR decisions to complete frontend Design Doc approval
Design Doc compliance and security validation with optional auto-fixes
Showing SKILL.md
| name | recipe-fullstack-build |
| description | Execute materialized fullstack task files with layer-aware agent routing |
| disable-model-invocation | true |
Execute Skill: llm-friendly-context before writing Agent prompts, handoffs, or generated artifacts. Execute Skill: subagents-orchestration-guide before making workflow decisions, invoking agents, or resolving findings.
Core Identity: "I am an orchestrator." (see subagents-orchestration-guide skill)
Local authority gate: Make this recipe's workflow decisions and validate each returned result directly; delegate semantic deliverable production to the named specialist.
Review Resolution Gate [MANDATORY]: Resolve every actionable deliverable-review finding through subagents-orchestration-guide Review Resolution before correction or progression.
Before the first finding disposition, read references/review-resolution.md from the loaded subagents-orchestration-guide skill.
MANDATORY: Read references/monorepo-flow.md from subagents-orchestration-guide skill BEFORE proceeding. Follow the Extended Task Cycle and Agent Routing defined there.
*-backend-task-* → task-executor + quality-fixer*-frontend-task-* → task-executor-frontend + quality-fixer-frontendCRITICAL: Run layer-appropriate quality-fixer(s) before every commit.
Work plan: $ARGUMENTS
Before any task processing, locate the work plan. Resolution rule:
$ARGUMENTS when present.{plan-name}-{backend|frontend}-task-*.md naming contract and map each group to docs/plans/{plan-name}.md.docs/plans/ when exactly one exists.Compute the Consumed Task Set for this run — the exact files this recipe owns, executes, and later deletes. Use the same restricted pattern as Work Plan Resolution:
docs/plans/tasks/ matching the layer-aware patterns {plan-name}-backend-task-*.md and {plan-name}-frontend-task-*.md for the {plan-name} resolved by Work Plan Resolution. Single-layer tasks are excludedEvery subsequent reference to "task files" in this recipe — Task Generation Decision Flow, Task Execution Cycle iteration, and Final Cleanup — uses this set, not the unrestricted docs/plans/tasks/*.md glob.
Analyze the Consumed Task Set and determine the action required:
| State | Criteria | Next Action |
|---|---|---|
| Tasks exist | Consumed Task Set is non-empty | User's execution instruction serves as batch approval → Enter autonomous execution immediately |
| No tasks + approved plan exists | Consumed Task Set is empty but the resolved work plan has batch approval | Run task-decomposer; the approval already authorizes mechanical task materialization |
| No tasks + unapproved plan exists | Consumed Task Set is empty and the resolved work plan is not approved | Review it when needed, then present the plan approval gate before task materialization |
| Neither exists + Design Doc exists | No plan, no Consumed Task Set, but docs/design/*.md exists | Invoke work-planner to create a work plan, then run document-reviewer (dev-workflows:document-reviewer, doc_type: WorkPlan). Run Review Resolution through its correction re-review, escalation, and convergence transitions, using work-planner for rerouted corrections; then present the resolved plan for batch approval before task materialization |
| Neither exists | No plan, no Consumed Task Set, no Design Doc | Report missing prerequisites to user and stop |
When the Consumed Task Set is empty:
Use the normal Work Plan review and approval gate when batch approval is absent. Existing batch approval authorizes task materialization directly.
Invoke task-decomposer using Agent tool:
subagent_type: "dev-workflows:task-decomposer"description: "Materialize work plan tasks"prompt: "Read work plan at docs/plans/[plan-name].md and output individual single-commit task files in docs/plans/tasks/. Use layer-aware naming: {plan}-backend-task-{n}.md, {plan}-frontend-task-{n}.md from each Work Plan task's Executor lane."Recompute the Consumed Task Set using the same restricted pattern from the Consumed Task Set section above. When it remains empty, apply Specialist Result Acceptance: validate the invocation and returned artifacts, correct recoverable input or naming errors, and rerun. Stop for the user only when resolving the plan or intended layer boundary requires a user-owned decision.
MANDATORY: For each task in the Consumed Task Set, route agents by task filename pattern from monorepo-flow.md reference.
| Filename Pattern | Executor | Quality Fixer |
|---|---|---|
*-backend-task-* | dev-workflows:task-executor | dev-workflows:quality-fixer |
*-frontend-task-* | dev-workflows-frontend:task-executor-frontend | dev-workflows-frontend:quality-fixer-frontend |
MANDATORY EXECUTION CYCLE: execute → branch on executor result → quality-fix → commit
Before the loop, register "Execute consumed task set", "Run post-implementation verification", "Clean up consumed task files", and "Report completion" once with TaskCreate; mark and advance the active phase with TaskUpdate.
For EACH task, YOU MUST:
diffBase, pass task_file: [path], and receive the structured responsestatus: "escalation_needed" or "blocked" → Apply subagents-orchestration-guide Specialist Result Acceptance; escalate only a valid user-owned blockrequiresTestReview is true → Identify the changed integration/E2E test files in the current changes and invoke integration-test-reviewer with them as changedTestFiles, plus diffBase, taskFile, prompt-only claims, and mutationEvidence
approved → Proceed to step 3blocked → Apply Specialist Result Acceptanceneeds_revision → Pass qualityIssues unchanged into the Review Resolution Gate; return to step 1 for rerouted corrections and derive convergence from correction re-review prior_feedback_reconciliationstatus: completed → Proceed to step 3task_file, upstream mutationEvidence, and qualityCommand when available (caller first, otherwise current task)
stub_detected → Return to step 1 with the layer quality-fixer's incompleteImplementations array unchanged as the canonical incompleteImplementations fieldblocked → Apply Specialist Result Acceptanceverification_incomplete → Retain the complete result for final retry and proceed to step 4approved → Proceed to step 4approved or verification_incomplete; append its verification trailers for the latterUse each subagent's semantic result and repository evidence through Specialist Result Acceptance; canonical status fields provide the normal routing shortcut. Proceed to the next task after step 4 and retain any verification limitation with its status kept proof-limited.
Verify task files exist per Pre-execution Checklist, then enter autonomous execution mode. When requirement changes are detected during execution, escalate to the user with the change summary before continuing.
Before invoking post-implementation verifiers, apply subagents-orchestration-guide's retained verification limitation retry with each layer's quality-fixer. Continue with the verifiers after clearing or retaining each result; include only repeated limitations in the completion report.
Resolve all readable Design Docs from the Work Plan, or the Work Plan itself when none exist; missing input blocks verification.
Emit one code-verifier call per resolved document plus one security-reviewer call in one assistant message, then await all:
doc_type and single document_pathgoverningDocuments listApply subagents-orchestration-guide's Post-Implementation Verification status-routing and fix/re-run rules with the layer-appropriate executor and quality-fixer. Present the unified report; proceed to Final Cleanup after the complete verification set reaches Review Resolution convergence.
Before the completion report, delete the implementation task files this recipe consumed. Their work is committed; docs/plans/ is ephemeral working state and is not retained between recipe runs:
docs/plans/{plan-name}.md) — the user decides whether to delete it after final reviewIf task-file deletion fails with a filesystem error, report the failure and continue to the completion report.
Final report must include: