一键导入
run-blueprint
Execute all tasks in a blueprint using build-task workflow
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
菜单
Execute all tasks in a blueprint using build-task workflow
用 Codex 或 Claude 帮你安装 复制这段 Prompt,粘贴到 Codex、Claude 或其他助手里,让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | run-blueprint |
| description | Execute all tasks in a blueprint using build-task workflow |
| argument-hint | [blueprint-file.md] |
Execute all tasks in a blueprint plan sequentially, respecting dependencies.
plan_path: $ARGUMENTS (required)
TASK_TRACKING = config_read("task_tracking.enabled", "false")
If TASK_TRACKING:
BLUEPRINT_ID = "run-blueprint-{timestamp}"
TASK_MAP = {} # Maps T1, T2 etc to Claude Task IDs
Read(file_path: plan_path)
Parse the ## Implementation Tasks section:
| ID | Task | Points | Dependencies | Status |
|---|---|---|---|---|
| T1 | Setup auth middleware | 2 | - | ⏳ |
| T2 | Create login endpoint | 2 | T1 | ⏳ |
Task status icons:
If TASK_TRACKING:
For each ROW in implementation_tasks:
TASK = TaskCreate:
subject: "{ROW.id}: {ROW.task}"
activeForm: "Building {ROW.task}"
metadata: {blueprint_id: BLUEPRINT_ID, task_id: ROW.id, points: ROW.points, source: plan_path}
TASK_MAP[ROW.id] = TASK.id
# Set up dependencies
For each ROW in implementation_tasks where ROW.dependencies != "-":
BLOCKED_BY = [TASK_MAP[dep] for dep in ROW.dependencies.split(",")]
TaskUpdate(TASK_MAP[ROW.id], addBlockedBy: BLOCKED_BY)
If TASK_TRACKING:
TASKS = TaskList()
BLUEPRINT_TASKS = filter(TASKS, task.metadata.blueprint_id == BLUEPRINT_ID)
NEXT_TASK = first(BLUEPRINT_TASKS where status == "pending" AND blockedBy.length == 0)
Else:
For each task in Implementation Tasks:
If task.status == ⏳:
If task.dependencies all ✅:
NEXT_TASK = task
Break
If no task found (all ✅ or 🔴): Go to Step 5.
Skill("build-task") "{task.id}" --no-ship
If TASK_TRACKING:
CLAUDE_TASK_ID = TASK_MAP[NEXT_TASK.id]
TaskUpdate(CLAUDE_TASK_ID, status: "in_progress")
# After build completes
If build succeeded: TaskUpdate(CLAUDE_TASK_ID, status: "completed")
If build failed: TaskUpdate(CLAUDE_TASK_ID, status: "completed", metadata: {result: "failed"})
# Always update markdown (for visibility in file)
Edit(file_path: plan_path):
If build succeeded: task.status = ✅
If build failed: task.status = 🔴
Go to Step 2 (next task).
When all tasks are ✅ or 🔴:
Skill("ship-it")
| System | Format | Example |
|---|---|---|
| GitHub Issues | #{number} | #123 |
| Linear | {PROJECT}-{number} | PROJ-123 |
| Beads | BEADS-{number} | BEADS-123 |
| File-based | TODO-{number} | TODO-123 |
| Scenario | Action |
|---|---|
| Task build fails | Mark 🔴, continue to next task |
| All dependencies failed | Skip task, mark 🔴 |
| Blueprint file not found | Error, exit |
| No tasks section found | Error, exit |
Create distinctive, production-grade frontend interfaces with high design quality. Use when building web components, pages, or applications. Includes framework-specific guidance for Tailwind, React, Vue, and Rails/Hotwire ecosystems.
Skill file structure, naming conventions, directory layout, frontmatter requirements, and invocation control. Use when creating skill files or slash commands to ensure correct format and validation.
Forces adversarial reasoning before committing to decisions. Triggers on architectural choices, approach selection, and planning phases to prevent premature commitment bias.
Verify LLM reasoning using Reverse Chain-of-Thought (RCoT) to detect overlooked conditions.
Autonomous task implementation - research, plan, build, review, fix, ship
Generate a complete favicon set (ICO, PNG variants, apple-touch-icon, web manifest) from a source image and integrate into the project's HTML layout. Use when user asks to generate favicons, set up PWA icons, or add an apple-touch-icon.