List Open Work logo

List Open Work

Community
OutThisLife
list-open-work

List my open MRs/PRs in the current repo/worktree, each with its tracker ticket. Use when I invoke /list-open-work, or ask what I have open, what's in review, or for a standup list of my open MRs or PRs.

Overview

PublisherOutThisLife
Repositorybrooklyn-skills
Skill namelist-open-work
Stars
200
Forks
6
Bundled files
Instructions only
LicenseMIT
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • Self-contained

    Everything the model needs lives in the instructions — no extra files to sync.

  • Open source

    Published by OutThisLife on GitHub. Read the source before you install it.

Installation

Install the List Open Work AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/OutThisLife/brooklyn-skills/tree/main/skills/list-open-work
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/OutThisLife/brooklyn-skills.git /tmp/brooklyn-skills
mkdir -p .claude/skills
cp -r /tmp/brooklyn-skills/skills/list-open-work .claude/skills/list-open-work
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable List Open Work in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads List Open Work on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so List Open Work is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

/list-open-work

Answer one question: what of mine is open in this repo, and which ticket does each carry? Lead with the list. Anything else goes after it, in prose.

This repo only

Scope is the git checkout you are in (worktree or primary). Read its origin. List my open MRs/PRs against that remote. Stop there.

  • Not a git repo → say so and stop. Do not hunt another project.
  • Do not list another origin, org, forge, or tracker from memory, another session, or another worktree.
  • Infer the ticket key from this repo's branches/titles (e.g. PROJ-123 in feat/PROJ-123-slug). Never reuse a prefix you saw somewhere else.
bash
git rev-parse --show-toplevel
git remote get-url origin

Output format

Write it straight into the chat as ordinary rendered markdown. Never wrap it in a code fence — the links have to be clickable.

Blank line between every item. Markdown treats a single newline as a space, so adjacent lines become one paragraph. That is a failure. Each MR is its own paragraph: newline, blank line, next line.

Exactly this shape. Header is Open MRs: or Open PRs:. No bullets, no emoji. The number and the ticket key are both links. Ticket sits in trailing [KEY].

Open MRs:

!42 - shorten the save path [PROJ-101]

!41 - default the new flag on [PROJ-99]

!30 - expose status history on the public API

  • Descending by number. !N on GitLab, #N on GitHub. "MRs" vs "PRs" likewise.
  • One paragraph each: <!|#><N> - <title> [<KEY>]. No bold.
  • Ticket in trailing [KEY]. Move it there if the author buried it mid-title: perf: PROJ-77 — cut over-fetch becomes perf: cut over-fetch [PROJ-77].
  • No ticket → omit the [KEY]. Don't invent one, don't write (no ticket).
  • Keep the MR/PR title as-is (minus the relocated key). Don't rewrite it.

Wrong (renders as one wrapped line):

Open MRs:
!42 - shorten the save path [PROJ-101]
!41 - default the new flag on [PROJ-99]

Pasting into Slack is a separate ask. Forge/tracker integrations linkify !42 and PROJ-101 on their own, so that version is the same lines with the link syntax stripped — still one item per line.

Collect

Mine means authored by me. Resolve my username from the API — never guess it.

GitLab origin:

bash
glab api user
glab api "projects/<url-encoded-path>/merge_requests?state=opened&scope=all&per_page=100"

Filter to author = me. <url-encoded-path> is this origin's path, not a path from another repo.

GitHub origin: gh pr list --author @me --state open --json number,title,headRefName,body

Ticket = first \b[A-Z][A-Z0-9]+-\d+\b in the title, then the branch, then the description. Use that key's tracker host as this repo uses it — don't guess a site from another company.

Jira account verification

Validate actual Jira results against the repository's tracker project; acli jira auth status alone can report the intended site while searches still reach another account. If project keys or API hosts contradict the expected workspace, explicitly select the saved account with acli jira auth switch --site <repo-tracker-host> --email <verified-email>, then read back a known ticket before trusting the list. Do not reuse results from the wrong site.

Use acli jira workitem search --paginate and compare the distinct fetched count against the same JQL with --count. Search has a restricted field list; retrieve timestamps, descriptions, and comments with workitem view --fields summary,description,status,priority,assignee,updated,comment,issuelinks --json instead. Read comments before classifying Testing tickets as new implementation work.

Then, in prose

Only the things that change what I'd do next:

  • Conflicts with the default branch — blocked, not in review. GitLab recomputes mergeability lazily, so a plain MR GET happily returns a stale has_conflicts: false. Force it and read merge_status:
bash
glab api "projects/<path>/merge_requests/<iid>?include_diverged_commits_count=true"
# cannot_be_merged + has_conflicts → needs a rebase; diverged_commits_count = how far behind
  • Red pipelines (head_pipeline.status).
  • Ticket status that contradicts the MR: green with every thread resolved but the ticket still says In Progress.
  • No ticket plus months of age — give the age, offer to close.

Skip approvals unless asked. Don't narrate fields that changed nothing.

For GitHub readiness checks, read submitted review bodies and human PR comments as well as paginated inline threads. A COMMENTED review can contain explicit merge blockers even when unresolved inline-thread count is zero and CI is green. Match each review to its reviewed head; older findings need rechecking rather than being asserted as current defects.

Keep CI conclusions, conflict status, and substantive review blockers separate. Query gh pr checks for each current head; UNKNOWN mergeability is unverified, not conflict-free. Persist the inventory and aggregate counts programmatically before reporting totals.

When this repository uses GitHub issues rather than tracker keys, inspect explicit issue links in the PR body as well as closingIssuesReferences. An intentionally non-closing reference can be the real attached ticket. Read the issue's state and discussion before changing it; a partial fix or pending live verification can require keeping it open.

Frequently asked questions

What does the List Open Work AI skill do?

List my open MRs/PRs in the current repo/worktree, each with its tracker ticket. Use when I invoke /list-open-work, or ask what I have open, what's in review, or for a standup list of my open MRs or PRs.

Why use List Open Work on TypingMind?

Because you install it once and use it with any model. List Open Work is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install List Open Work in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/OutThisLife/brooklyn-skills/tree/main/skills/list-open-work. TypingMind reads its SKILL.md and installs it as a skill you can enable per chat.

Which AI models can use List Open Work?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with List Open Work?

As many as you like. As long as a model supports skills, you can use List Open Work with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the List Open Work AI skill free?

Yes. It is published on GitHub by OutThisLife under the MIT license. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇