to-tickets

$npx mdskill add JasonxzWen/harness-hub/to-tickets

Decompose accepted specs into dependency-aware, project-owned implementation slices.

  • Converts accepted specifications into independently verifiable execution slices.
  • Writes into existing project task conventions or local fallback state.
  • Decomposes vertically with tracer bullets that prove architecture.
  • Returns proposed decomposition to main agent without creating storage.

SKILL.md

.github/skills/to-ticketsView on GitHub ↗
---
name: to-tickets
description: Load when an accepted specification must be decomposed into dependency-aware, project-owned implementation slices without creating a new tracker or publishing remote issues by default.
license: MIT
metadata:
  source: "mattpocock/skills skills/engineering/to-tickets"
  upstream_commit: "e9fcdf95b402d360f90f1db8d776d5dd450f9234"
---

# To Tickets

Convert an accepted specification into the fewest independently verifiable execution slices the project can actually own. This Skill decomposes work; it does not become a task tracker, scheduler, or implementation workflow.

## Choose The Existing Owner

Write into the existing project task, issue, or plan convention when one exists and the user authorized local mutation. If no durable project convention exists, use the existing `.harness-hub/state/current-task.md` as the local fallback when available.

If neither exists, return the proposed decomposition to the main Agent without creating a storage system. Do not create `.harness-hub/tasks/`, a new registry, or hidden run state.

Remote publication requires explicit authorization. Do not create, label, assign, comment on, or close remote issues merely because a decomposition is ready.

## Decompose Vertically

Start with one vertical tracer-bullet that crosses the real user or caller path and proves the architecture. Each subsequent slice must deliver observable behavior or unlock a named dependent slice.

For every slice record:

- outcome and accepted spec anchors
- included behavior and explicit non-goals
- touched public seam or integration path
- deterministic acceptance evidence
- blocking edges to other slices
- risks or decisions that prevent safe execution

Avoid layer tickets such as “build database,” “build API,” and “build UI” when none is independently useful or verifiable.

## Order By Dependency

Build only the dependency graph needed for execution. The current frontier is the set of slices whose blocking edges are already resolved. Prefer a wide frontier of independent slices after the tracer bullet rather than a false serial chain.

Use expand/contract only when a broad refactor genuinely requires a compatibility window; otherwise change the direct path once. Do not invent adapters or migration stages for hypothetical consumers.

## Handoff

Return:

```text
TRACER BULLET
- smallest end-to-end proof

CURRENT FRONTIER
- independently executable slices

WAITING
- slice -> unresolved blocking edge

ACCEPTANCE
- exact project-defined evidence per slice
```

The main Agent owns slice implementation, integration, and user reporting. Native Subagents may perform only bounded, independent, read-only investigation, review, or verification under the Host and project contracts.

More from JasonxzWen/harness-hub

SkillDescription
agent-interaction-auditLoad when the user asks for private agent interaction audits from local Codex/Claude Code traces, task profiles, tool-call decisions, collaboration bottlenecks, prompt/rule or automation-log audits, closeout retrospectives, or evidence-backed Skill/project-rule/Eval/SOP/OKF improvements.
agent-reachLoad when a task needs read-only internet research or content retrieval through an already installed Agent Reach CLI; diagnose available backends first, never install, upgrade, authenticate, configure, or copy credentials on the user's behalf.
animation-vocabularyLoad when a user describes a Web animation or motion effect without knowing its name and needs the precise term or close alternatives; use only for naming, not design, implementation, or review.
apple-designLoad when a frontend task needs Apple-platform design guidance or physical, gesture-driven, velocity-aware, interruptible interaction; skip generic visual styling and use review-animations for report-only motion review.
beautiful-feishu-whiteboardLoad when creating or restyling a polished editable Feishu or Lark whiteboard, infographic, diagram, poster, or visual explainer; do not load for generic SVGs, static images, or other whiteboard platforms.
clone-websiteLoad when the user explicitly asks to clone, reverse-engineer, or rebuild a website they own or are authorized to reproduce; do not load for generic redesign, inspiration, scraping, phishing, impersonation, or copying third-party brands.
codebase-to-courseLoad when a user wants a source-backed interactive HTML course that explains how a codebase works; do not load for chat tutoring, repository review, articles, slide decks, or general browser artifacts.
context-guidanceFixture skill used to represent an additional standard skill context surface.
decision-uiLoad when Codex 即将要求用户在 2-3 个互斥方案中做真实决定、准备询问阻塞性澄清问题或需要确认高影响取舍;将选择转换为宿主原生可点击 UI,并在工具不可用时诚实文本回退。不要用于低风险实现细节、只有一个合理答案的问题、普通信息采集或不需要用户决策的完成汇报。
design-taste-frontendLoad when a task needs anti-template frontend visual direction, design-read calibration, or pre-flight critique for landing pages, portfolios, marketing pages, or redesigns; do not load for dashboards, data tables, multi-step product UI, routine frontend logic, HTML reports, slide decks, or generic code explanation.