agentleFS
GavinGreenwoodGavinGreenwood/agentic-workflow-template/.agents/skills/pickup/SKILL.md

pickup

Skill5 starsChanged 44 days agoMITView on GitHub

User-invoked only. Assign the ticket, read it fully, brief the work, create the branch, start PROGRESS.md

  • Reads credentials

33 other projects use a version of this.

Adopted by

Most recently taken up by

Used with

Projects with it also use ai-review, discover and 10 more.

Other files people use alongside this one.

Files found in the same public projects as a version of this one, more often than they turn up anywhere else. Each project counts once, large collections count for less, and this file's own repository is left out.
  • SkyWalker2506Skill

    ai-review

    This skill should be used when the user asks to "review a PR", "code review", "AI review", "check the pull request", or asks about automated review via OpenRouter or Gemma. Also activates when setting up GitHub Actions for PR review.

    Also in 3 projects that use this

  • anombyte93Skill

    discover

    Phase 1 of the prd-taskmaster pipeline: brainstorm-driven discovery. Delegates to superpowers:brainstorming in Interactive Mode (one adaptive question at a time), or self-brainstorms in Autonomous Mode when no user is present. Intercepts before the brainstorming chain hands off to writing-plans — this skill owns the exit. Extracts constraints, calibrates scale (Solo / Team / Enterprise), and advances the pipeline to GENERATE.

    Also in 3 projects that use this

  • asterkinSkill

    adr

    Create Architecture Decision Records (ADRs) following project standards. Guides template usage, process, and best practices for documenting architecturally significant decisions.

    Also in 3 projects that use this

  • AWSSkill

    retro

    Conduct a retrospective on a coding session.

    Also in 5 projects that use this

  • Hugging FaceSkill

    implement

    Make tests pass. Invoke after /write-tests produces failing tests.

    Also in 4 projects that use this

  • Microsoft AzureSkill

    design

    Design rules and conventions for the API Center Portal. Use when editing detail pages, card components, badges, layout, navigation, sidebars, or headers. Covers badge styling, metadata placement, page navigation patterns, and layout alignment.

    Also in 3 projects that use this

The file

Show its rules and text

We split this skill into the rules it gives an agent

24 rules. 0 are common, 9 are distinctive, and 15 are very specific to this project.

  • 0

    Common

    Most of the 29 edited versions we read keep these.

  • 9

    Distinctive

    This author's own choices. Worth a look.

  • 15

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Ticket input: use the value and flags supplied with the skill invocation.

    line 9
  2. Very specific:

    If --stay is present anywhere in the invocation input, enable stay mode and remove the flag before normalising.

    line 22
  3. Very specific:

    If --qa is present anywhere in the invocation input, enable QA-rework mode and remove the flag before normalising.

    line 23
  4. Very specific:

    If the remaining ticket ID is a bare number (e.g.

    line 25
  5. Very specific:

    Check that JIRA_BASE_URL, JIRA_API_TOKEN, JIRA_EMAIL, and JIRA_ACCOUNT_ID are set in .env.

    line 30
  6. Distinctive:

    Check the fetched ticket data:

    line 55
  7. Very specific:

    If fields.parent is present and its issuetype is "Epic", the epic is set → continue to Step 3.

    line 57
  8. Very specific:

    If fields.customfield_10014 is set (classic epic link field), the epic is set → continue to Step 3.

    line 58
  9. Distinctive:

    If neither is set → note this in the brief (Step 9) as an open question, then continue.

    line 59
  10. Very specific:

    If fields.assignee is null (unassigned) → proceed.

    line 63
  11. Very specific:

    If fields.assignee.accountId matches $JIRA_ACCOUNT_ID → proceed.

    line 64
  12. Very specific:

    If fields.assignee is someone else → STOP.

    line 65
  13. Distinctive:

    Do this before the In Progress transition (Step 6).

    line 78
  14. Distinctive:

    Read and summarise each subtask's summary, description, and status.

    line 123
  15. Very specific:

    If the ticket has a fields.parent, fetch that ticket too to understand the broader story or epic.

    line 127
  16. Very specific:

    If fields.customfield_10014 (epic link) is set and different from the parent, fetch that as well.

    line 129
  17. Distinctive:

    Run git branch --show-current to check the current branch.

    line 167
  18. Very specific:

    If the current branch is main (or empty/detached), STOP and tell the user: "You are on main.

    line 168
  19. Distinctive:

    If the merge reports conflicts, STOP and ask the user to resolve them before continuing.

    line 174
  20. Very specific:

    Every commit for this ticket must reference the new ticket ID in its message (e.g.

    line 175
  21. Distinctive:

    Use the ticket ID lowercased and a 2–4 word kebab-case description derived from the summary.

    line 184
  22. Very specific:

    Create a PROGRESS.md file in the repo root as the session scratchpad (per AGENTS.md).

    line 188
  23. Very specific:

    If stay mode (--stay) is NOT active: create PROGRESS.md fresh. If one already exists (leftover from a previous session), read it first, …

    line 213
  24. Distinctive:

    After the briefing is output, the branch is created (or confirmed, in stay mode), and PROGRESS.md is initialised (or appended to):

    line 252

“Common” is measured: kept by at least 6 in 10 of the 29 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

Pick up a Jira ticket: assign it to me, move it to In Progress, read it fully with all subtasks and linked context, then prepare to implement.

Ticket input: use the value and flags supplied with the skill invocation.

Usage: pickup <ticket-id> [--stay] [--qa]

  • --stay — do not create a new branch. Stay on the current branch and do the work there. Use this when stacking multiple tickets on one branch.
  • --qa — QA-rework mode. Work is already done and merged to main; the ticket is back from QA with comments. Runs the normal Jira front-door (assign, In Progress, board) — which qa-review-action doesn't — then hands off to qa-review-action for the feedback itself. Branch behaves like a normal pickup (fresh fix branch off main); add --stay to rework on the current branch with latest main merged in first.

Step 0 — Parse arguments and normalise the ticket ID

Split the invocation input into the ticket ID and any flags:

  • If --stay is present anywhere in the invocation input, enable stay mode and remove the flag before normalising. Stay mode changes Steps 10 and 11 only — all Jira steps run unchanged.
  • If --qa is present anywhere in the invocation input, enable QA-rework mode and remove the flag before normalising. QA-rework mode is described in Step 12 — all Jira steps (1–8) and the PROGRESS.md step (11) run unchanged; Step 9 adds one conditional brief line.
  • --stay and --qa can be combined. With both active, you do the QA rework on the current branch (stay mode) and Step 10 merges latest main into it first.
  • If the remaining ticket ID is a bare number (e.g. 25), prepend $JIRA_PROJECT_KEY- to get e.g. PROJ-25.
  • Store the cleaned, normalised ticket ID as TICKET_KEY. Use TICKET_KEY for all subsequent steps.

Step 1 — Verify credentials

Check that JIRA_BASE_URL, JIRA_API_TOKEN, JIRA_EMAIL, and JIRA_ACCOUNT_ID are set in .env. If any are missing, stop and ask the user to add them. Source .env before all Jira API calls.

Step 2 — Fetch the ticket

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  "$JIRA_BASE_URL/rest/api/3/issue/$TICKET_KEY?expand=subtasks,renderedFields"

Parse the response and extract:

  • fields.summary
  • fields.description (full rendered text)
  • fields.assignee (may be null)
  • fields.status.name
  • fields.issuetype.name
  • fields.parent (if present — epic or story)
  • fields.subtasks (array of subtask keys)
  • fields.issuelinks (linked issues)
  • fields.labels
  • fields.priority.name
  • fields.customfield_10014 (epic link, if present)

Step 2a — Check epic assignment

Check the fetched ticket data:

  • If fields.parent is present and its issuetype is "Epic", the epic is set → continue to Step 3.
  • If fields.customfield_10014 is set (classic epic link field), the epic is set → continue to Step 3.
  • If neither is set → note this in the brief (Step 9) as an open question, then continue.

Step 3 — Assignee safety check

  • If fields.assignee is null (unassigned) → proceed.
  • If fields.assignee.accountId matches $JIRA_ACCOUNT_ID → proceed.
  • If fields.assignee is someone else → STOP. Output: "$TICKET_KEY is assigned to [displayName] — are you sure you want to pick this up?" Do not proceed until the user confirms.

Step 4 — Assign the ticket to me

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  -X PUT "$JIRA_BASE_URL/rest/api/3/issue/$TICKET_KEY/assignee" \
  -H "Content-Type: application/json" \
  -d "{\"accountId\": \"$JIRA_ACCOUNT_ID\"}"

Step 5 — Move onto the active board

Do this before the In Progress transition (Step 6). Adding an issue to the board drops it into the board's default column (e.g. "Ready for Development"), which would overwrite an In Progress status set beforehand. Always add to the board first, then transition last.

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  -X POST "$JIRA_BASE_URL/rest/agile/1.0/board/$JIRA_BOARD_ID/issue" \
  -H "Content-Type: application/json" \
  -d "{\"issues\": [\"$TICKET_KEY\"]}"

Step 6 — Move to In Progress

This must be the last Jira write, so nothing (like the board-add above) can clobber the status afterwards.

First, get available transitions (fetch them fresh — available transitions and their ids depend on the ticket's current status, so never reuse an id from an earlier run):

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  "$JIRA_BASE_URL/rest/api/3/issue/$TICKET_KEY/transitions"

Find the transition whose name matches "In Progress" (case-insensitive). Use its id to apply it:

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  -X POST "$JIRA_BASE_URL/rest/api/3/issue/$TICKET_KEY/transitions" \
  -H "Content-Type: application/json" \
  -d '{"transition": {"id": "<transition-id>"}}'

Then confirm the status actually landed on In Progress (a 204 only means the request was accepted, not that the status stuck):

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  "$JIRA_BASE_URL/rest/api/3/issue/$TICKET_KEY?fields=status"

Step 7 — Fetch all subtasks

For each key in fields.subtasks, fetch the full issue:

source .env && curl -s -u "$JIRA_EMAIL:$JIRA_API_TOKEN" \
  "$JIRA_BASE_URL/rest/api/3/issue/<subtask-key>?expand=renderedFields"

Read and summarise each subtask's summary, description, and status.

Step 8 — Fetch parent/epic context

If the ticket has a fields.parent, fetch that ticket too to understand the broader story or epic.

If fields.customfield_10014 (epic link) is set and different from the parent, fetch that as well.

Step 9 — Output a brief

Print a structured briefing so the work is clear before any code is written:

## Ticket: $TICKET_KEY — <summary>

**Type:** <issuetype>
**Priority:** <priority>
**Status:** → In Progress (just moved)
**Parent:** <parent key + summary, if any>
**Branch (stay mode only):** <current branch — include this line only when `--stay` is active; omit otherwise>
**🔁 QA rework:** <only include this line when `--qa` is active — "Work is already merged to `main`. This is rework from QA comments. After branch setup, control hands off to `qa-review-action` to classify and action the feedback.">

### Description
<rendered description>

### Subtasks
- [ ] <subtask-key>: <summary> (<status>)
...

### Linked issues
- <link type>: <key> — <summary>
...

### Implementation notes
<what needs to be built, based on reading the ticket and relevant docs>

### Open questions
<anything unclear that needs resolving before coding starts>

Step 10 — Create the feature branch (skipped in stay mode)

If stay mode (--stay) is active:

  • Run git branch --show-current to check the current branch.
  • If the current branch is main (or empty/detached), STOP and tell the user: "You are on main. The --stay flag is for continuing work on an existing feature branch. Rerun without --stay to create a new branch, or switch to the feature branch first."
  • Otherwise: do not switch branches, do not checkout main, do not create a new branch. All work happens on the current branch.
  • If both --qa and --stay are active: bring the merged implementation into the current branch first so the QA fixes land on top of it:
    git fetch origin main && git merge origin/main
    If the merge reports conflicts, STOP and ask the user to resolve them before continuing. Do not proceed to the hand-off with an unfinished merge.
  • Every commit for this ticket must reference the new ticket ID in its message (e.g. $TICKET_KEY: feat: ...), even though the branch name references a different ticket.
  • Skip to Step 11.

Otherwise, following CONTRIBUTING.md naming conventions, create and switch to a new branch:

git checkout main && git pull && git checkout -b <ticket-id-lowercase>-<short-description>

Use the ticket ID lowercased and a 2–4 word kebab-case description derived from the summary.

Step 11 — Initialise PROGRESS.md

Create a PROGRESS.md file in the repo root as the session scratchpad (per AGENTS.md). Seed it with the ticket context so the session log starts with a clear baseline:

# PROGRESS.md — <ticket-id>: <summary>

**Branch:** <branch-name>
**Started:** <today's date>

## Plan

<high-level implementation approach derived from the ticket brief>

## Progress

_Nothing logged yet._

## Decisions

_None yet._

## Open Questions

<carry over any open questions from the Step 9 brief>

If stay mode (--stay) is NOT active: create PROGRESS.md fresh. If one already exists (leftover from a previous session), read it first, then overwrite it with the new session header — do not append to stale content.

If stay mode (--stay) is active and a PROGRESS.md already exists, it belongs to the branch's ongoing work — do not overwrite it. Instead, append a new ticket section to the end:

---

# <ticket-id>: <summary> (stacked on this branch)

**Branch:** <current branch>
**Started:** <today's date>

## Plan

<high-level implementation approach derived from the ticket brief>

## Progress

_Nothing logged yet._

## Decisions

_None yet._

## Open Questions

<carry over any open questions from the Step 9 brief>

If stay mode (--stay) is active and PROGRESS.md does NOT exist, create it fresh using the standard template above (same as non-stay mode).

Step 12 — Hand off to QA review (QA-rework mode only)

Only when --qa is active (otherwise skip). The Jira front-door (assign, In Progress, board) and branch are now set up — the state qa-review-action can't establish on its own.

Tell the user: "Ticket set up for QA rework — assigned, In Progress, on branch <branch-name>. Handing off to qa-review-action." Then read and follow .agents/skills/qa-review-action/SKILL.md for $TICKET_KEY to fetch, classify, and action the feedback. Don't duplicate that work here.


After the briefing is output, the branch is created (or confirmed, in stay mode), and PROGRESS.md is initialised (or appended to):

  • In QA-rework mode (--qa): proceed straight to Step 12 — do not pause for go-ahead, since qa-review-action has its own confirmation gates before applying fixes or posting to Jira.
  • Otherwise: pause and ask: "Ready to start — any questions before I begin?" Wait for the user's go-ahead before writing any code.

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.