agentleFS
pahlevikunpahlevikun/levistack/skills/delivery/create-tickets/SKILL.md

create-tickets

Skill0 starsChanged 19 days agoMITView on GitHub

Break a plan into thin end-to-end tickets that name what blocks them, quiz the user on the breakdown, then draft or create Jira stories and tasks with a lean Context/Goal/DoD template and plain-language prose via polyglot-copywriter. Use when breaking work into tickets, writing story text, pasting ticket markdown, or creating issues under an epic (markdown only, acli, or Atlassian MCP).

5 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

5 rules. 0 are common, 1 is distinctive, and 4 are very specific to this project.

  • 0

    Common

    Most of the 5 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ask only for missing facts (skip what the user or repo already gave):

    line 56
  2. Very specific:

    Do not invent epic keys, dates, or acceptance criteria.

    line 68
  3. Very specific:

    Use case id: create-jira-story (registry pack: section order and ticket tone).

    line 75
  4. Very specific:

    Always load simple-prose.md with the core pipeline.

    line 77
  5. Very specific:

    Follow references/jira-create.md: Atlassian MCP first, then acli, then REST, else fall back to Route A with a clear blocker message.

    line 97

“Common” is measured: kept by at least 6 in 10 of the 5 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Create tickets

Peer skill: polyglot-copywriter (simple vocabulary, create-jira-story use case, default register santai).

References: references/story-template.md, references/breakdown.md, references/jira-create.md.

Routing

Pick one route from intent. If unclear, ask once:

"What should I do with the Jira story?"
Options:
- "Markdown only (I'll paste into Jira)"
- "Create the ticket in Jira (MCP or acli)"
User saysRoute
paste, markdown only, draft, template, describe the story, no JiraA — Markdown only
create in Jira, file the ticket, acli, MCP, add to epic, push to backlogB — Create ticket

Route B still shows the markdown draft for sign-off when the work is non-trivial or the user has not pre-approved the text.

Clarify first

Ask only for missing facts (skip what the user or repo already gave):

TopicWhy
Output routeA vs B (table above)
Source materialSpec, doc, MR, chat, or codebase area
CountOne story vs breakdown into several
Project key and issue typeStory, Task, Bug, etc.
Parent epic / linkRequired for B when the team uses epics
LanguageDefault English; Indonesian or other if they ask
Labels, componentsOnly if the team uses them

Do not invent epic keys, dates, or acceptance criteria. Mark unknown facts [TK] and ask.

Workflow (both routes)

  1. Break down — If the source is large, propose a story list (title, one line, blocked by) in thin end-to-end slices, then quiz the user on it and change it until they agree: breakdown.md. Do not write full bodies or publish before they say the list is right. If the source is a sliced plan (see write-plan, references/slicing.md), keep one story per slice and carry each slice's Blocked by over. Do not invent blockers when the source has none.
  2. Draft structure — Fill story-template.md. Keep sections lean; merge fluff into Context or drop it.
  3. Polyglot pass — Load polyglot-copywriter in rewrite mode:
    • Use case id: create-jira-story (registry pack: section order and ticket tone).
    • Register: santai unless the user asked for formal (profesional / baku).
    • Always load simple-prose.md with the core pipeline. Run the humanize pass on section prose, not on code paths or issue keys.
  4. Deliver — Route A or B below.

Route A — Markdown only

Emit a single copy-paste ready block per story:

### Summary
<imperative title line>

### Description
<full body with ## section headings from the template>

For multiple stories, separate blocks with a horizontal rule --- or numbered headings. When a story is blocked by another, add a ### Blocked by line under the description that names the blocking story by its summary, or "None, can start now". List stories in dependency order, blockers first. Do not call Jira. Tell the user they can paste the Description into Jira (wiki/markdown editor depending on site settings).

Route B — Create ticket

  1. Confirm project, type, parent epic (if any), and summary lines.
  2. Follow references/jira-create.md: Atlassian MCP first, then acli, then REST, else fall back to Route A with a clear blocker message.
  3. Blocked stories: create the stories in dependency order (blockers first) so each one can point at a real key. Then link them: each blocked story is blocked by its blockers. Use the Atlassian MCP's link tool if it has one. Otherwise follow references/jira-create.md (links section). Never put a placeholder key in a link.
  4. Return issue key(s), summary, and browse URL when the site is known.

Checklist before done

  • A multi-ticket breakdown was quizzed with the user before any body was written.
  • Every DoD bullet is testable (QA or reviewer can verify without guessing).
  • Prose passed polyglot simple-vocab rules (short sentences, no AI tell closers).
  • No secrets, employer-only names, or personal paths in ticket text.
  • Route A: user received paste-ready markdown. Route B: create confirmed or fallback explained.

Related skills

  • write-plan: the work is not broken down yet. Its slicing rules give each story its blockers.
  • challenge-me: stress-test the scope before you write tickets.
  • polyglot-copywriter: polish the story text.
  • standup: report ticket movement in the daily update.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.