一键导入
speckit-ultraplan-tasks
Generate a deeply-reasoned, implementation-ready task breakdown for the current spec-kit feature using extended analysis.
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
Generate a deeply-reasoned, implementation-ready task breakdown for the current spec-kit feature using extended analysis.
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
Review a production UX checkout harness run against the deployed Studio (daikonic). Use after `pnpm --filter @rune-langium/studio run test:prod-ux` (or test:prod-smoke) completes, when the user asks to review/triage a prod checkout run, or when a nightly prod-ux artifact needs auditing. Consumes report/run-manifest.json + screenshots + traces and produces a severity-ranked UX review with per-journey verdicts.
Documentation site and generated agent skills for rune-langium APIs
Documentation site and generated agent skills for rune-langium APIs Use when: Validating a single `.rosetta` file or snippet in memory.
Documentation site and generated agent skills for rune-langium APIs Use when: Embedding a Rune DSL language server in a web application via WebSocket.
Documentation site and generated agent skills for rune-langium APIs Use when: Rendering two or more `RuneTypeGraph` components simultaneously (different....
Runtime <ZodForm> renderer for Zod v4 schemas Use when: You need form rendering in storybook, playgrounds, or low-traffic admin UIs —.... Also: zod, zod-v4, react, forms, form-generation, react-hook-form, schema-driven, dynamic-forms, form-renderer, hookform-resolver, zod-form-renderer.
| name | speckit-ultraplan-tasks |
| description | Generate a deeply-reasoned, implementation-ready task breakdown for the current spec-kit feature using extended analysis. |
| compatibility | Requires spec-kit project structure with .specify/ directory |
| metadata | {"author":"github-spec-kit","source":"workflows:commands/ultraplan-tasks.md"} |
ultrathink
You are generating a comprehensive task breakdown for the current spec-kit feature by deeply analysing the specification and plan.
cd "$(git rev-parse --show-toplevel)" && \
source .specify/scripts/bash/common.sh && \
get_feature_paths
This provides FEATURE_DIR, FEATURE_SPEC (spec.md), IMPL_PLAN (plan.md), and TASKS (tasks.md). Load and carefully read all three files before continuing.
Conduct a thorough analysis of the specification and plan:
Overwrite $TASKS (tasks.md) with the full breakdown using this format:
# Tasks: [Feature Name]
> Generated by speckit.ultraplan — edit freely.
<!-- Clarifications (remove when resolved):
- [Any ambiguities spotted during analysis]
-->
## Implementation Tasks
- [ ] T001: [Concrete, single-concern task — one commit]
- [ ] T002: [Concrete, single-concern task — one commit]
...
Rules:
After writing tasks.md, output:
Ultraplan complete
**Feature**: [feature name]
**Tasks generated**: [N]
**Tasks file**: [absolute path to tasks.md]
### Task summary
T001 — [one-line description]
T002 — [one-line description]
...