explaingit

aaddrick/ticketmill

Analysis updated 2026-05-18

5JavaScriptAudience · developerComplexity · 4/5Setup · hard

TLDR

A Claude Code plugin that autonomously works through a batch of GitHub issues with built-in review, testing, and a human-approved final pull request.

Mindmap

mindmap
  root((ticketmill))
    What it does
      Batch processes issues
      Research and planning
      Opens one batch pull request
    Tech stack
      JavaScript
      Claude Code workflow
      Git worktrees
      GitHub CLI
    Safety features
      Contrarian gate
      Tests never skipped
      Circuit breakers
      Resumable runs
    Use cases
      Batch fix issues
      Select by label
      Dry run preview
    Requirements
      Authenticated gh CLI
      Per repo profile
      Working test command

Code map

Detail Auto

An interactive map of this repo's files and how they connect — its source is parsed live in your browser. Click Visualize to build it.

filefunction / class

What do people build with it?

USE CASE 1

Batch-process a list of GitHub issue numbers into one reviewed pull request.

USE CASE 2

Select issues to work on automatically by label instead of listing numbers by hand.

USE CASE 3

Resume an interrupted batch run without losing completed work.

USE CASE 4

Get a dry-run preview of which issues would be skipped, reviewed, or implemented before committing to a real run.

What is it built with?

JavaScriptClaude CodeGitGitHub CLI

How does it compare?

aaddrick/ticketmill00kaku/wp-rest-playgroundartefactdesigntn/jebari-imagebatch
Stars555
LanguageJavaScriptJavaScriptJavaScript
Setup difficultyhardhardeasy
Complexity4/53/52/5
Audiencedeveloperdevelopergeneral

Figures from each repo's GitHub metadata at analysis time.

How do you get it running?

Difficulty · hard Time to first run · 1h+

Requires an authenticated GitHub CLI with write access, git worktree support, and a proven working install and test setup for the target repo.

In plain English

Ticketmill is a Claude Code plugin that processes a batch of GitHub issues on its own, from research through to a pull request, while building in a lot of checks along the way. For each issue you point it at, it runs through research, an approach that gets challenged by a skeptical review step, a task plan that also gets challenged, implementation of each task with review and quality checks, a test run, and creation of a pull request. All the individual issue pull requests are merged into one batch branch, and the run finishes by opening a single combined pull request that a human still has to review and merge. The tool never merges anything into your main branch by itself. The project is designed to work with any technology stack. It learns your project's tools and testing commands from a settings file you set up once per repository, and it uses your project's own defined helper agents, creating new ones if some are missing. Before it will run at all, it needs Claude Code with a certain workflow feature, the GitHub command line tool logged in with write access, git with support for multiple working copies, and a proven, working setup for installing dependencies and running tests. A lot of the design is aimed at trust and safety. Tests are never quietly skipped, and if they are turned off on purpose that choice is clearly flagged for the human reviewer. Before any code is written, the proposed approach and plan are challenged by a built in skeptic. The tool keeps track of which issues are already claimed so that two people running batches at the same time do not duplicate work. If a pull request runs into a conflict after review, it attempts one automatic fix and retest before giving up and asking a person to step in. If too many issues fail, the whole run stops and gives you a plan to pick back up rather than continuing to burn through work. Every step of a run is posted as a comment on the relevant GitHub issue or pull request, and detailed logs are written to a folder in your project so you can review exactly what happened.

Copy-paste prompts

Prompt 1
Help me set up ticketmill in my repo by running the mill-init onboarding step.
Prompt 2
Explain what the contrarian gate does before ticketmill starts writing code for an issue.
Prompt 3
Show me how to run ticketmill against a set of specific GitHub issue numbers targeting my dev branch.
Prompt 4
Walk me through what happens if a ticketmill batch run hits a merge conflict on one of its pull requests.

Frequently asked questions

What is ticketmill?

A Claude Code plugin that autonomously works through a batch of GitHub issues with built-in review, testing, and a human-approved final pull request.

What language is ticketmill written in?

Mainly JavaScript. The stack also includes JavaScript, Claude Code, Git.

How hard is ticketmill to set up?

Setup difficulty is rated hard, with roughly 1h+ to a first successful run.

Who is ticketmill for?

Mainly developer.

Open on GitHub → Explain another repo

This repo across BitVibe Labs

Verify against the repo before relying on details.