Issuer lets you define all your GitHub Issues work tickets in one place as YAML, apply defaults, and post them to GitHub Issues in bulk.
-
Bulk issue creation from a single YAML file
-
Dry-run mode for testing without creating issues
-
Automatic milestone creation if milestones don’t exist
-
Configurable defaults and label application
-
Environment variable support for authentication
-
Issue validation with helpful error messages
-
GitHub API integration
Future plans include extending this capability to Jira (#9), GitLab Issues (#33), GitHub Projects, and other services.
This tool is oriented as much for project managers as it is for software developers, but the target audience is probably people who manage software development cycles, especially using Git tools and GitHub Issues (or soon Jira Issues and GitLab Issues).
📎
|
Issuer does not require Git be installed on your local workstation. The only direct use of Git would be if you commit and share the batch-configuration file with your team for collective editing. |
This tool is great at creating “stub” issues for entire teams, based on user-story documents, task lists, or other planning documents.
Once you convert such documents to our special YAML-based definition format (IMYML), use issuer
to post them to your GitHub Issues system via GitHub’s API.
Individual issue entries can be as basic or as detailed as you wish. They may consist of as little as just a title (summary) line, or that plus an assignee responsible for filling out the rest and adding metadata. Or they can be complete with labels, milestones, a type designation, and a complete explanation of the issue.
For end users, there are two ways to execute issuer
.
If you do not already operate in a Ruby environment, the Docker approach is recommended.
If you are not already a Ruby user, the issuer
utility may be best used from our Docker image.
📎
|
You will need Docker Desktop installed directly on MacOS or with WSL2 backend on Windows, but all instructions here work in the terminal. For Linux, use the Docker Engine install docs if you’re not already using Docker. |
With Docker installed and running…
-
Download the Docker image.
docker pull docopslab/issuer
-
Run the
issuer
command via a container based on the Docker image.docker run -it --rm -v $(pwd):/workdir -e ISSUER_API_TOKEN=$GITHUB_TOKEN docopslab/issuer --version
For actual issue creation, use something like:
docker run -it --rm -v $(pwd):/workdir -e ISSUER_API_TOKEN=$GITHUB_TOKEN docopslab/issuer your-issues.yml
The above command:
-
mounts your local directory to be readable by the Docker container
-
passes your GitHub access token environment variable into the container (
$ISSUER_API_TOKEN
could be$ISSUER_GITHUB_TOKEN
,$GITHUB_ACCESS_TOKEN
, or$GITHUB_TOKEN
). -
connects to GitHub and creates issue entries one at a time from
your-issues.yml
Everything after docopslab/issuer
accepts the standard arguments and options of the issuer CLI.
See Usage for more.
Optionally alias the base Docker command.
In your shell configuration (usually ~/.bashrc
or `~/.zshrc), add the following:
alias issuer='docker run -it --rm -v $(pwd):/workdir -e ISSUER_API_TOKEN=$GITHUB_TOKEN docopslab/issuer'
Reload your shell configuration for the alias to take effect:
source ~/.bashrc
If you have Ruby on your workstation, there are two common ways to install this gem.
gem install issuer
Then you can use the issuer
command from anywhere in your system.
Once installed, you can start using issuer
to create issues in GitHub.
The following steps assume the gem is either installed globally or issuer
is established as an alias (Docker method).
For Ruby Bundler usage, prepend bundle exec ` and for un-aliased Docker usage, prepend `docker run -it --rm -v $(pwd):/workdir -e ISSUER_API_TOKEN=$GITHUB_TOKEN docopslab/issuer
.
-
Prepare your issue definitions in an IMYML file (see examples and docs below).
-
Perform a “dry run” to validate your file and check what would be posted (no GitHub API calls made):
issuer example.yml --dry
-
Establish a Personal Access Token for GitHub and store it as an environment variable (see Authentication below).
-
Post issues to GitHub:
issuer example.yml
The specially formatted configuration files are structured as IMYML, for Issue Management YAML-based Modeling Format.
The YAML file must have the following structure:
$meta: # optional block for establishing general modes/settings
proj: org/repo
defaults: # value to infer when given property missing
vrsn: 0.1.0 # milestone/version
user: alice # assigned user
type: Bug # type of issue (must already be registered)
tags: [needs:labels, +posted_by_issuer, +needs:docs] # labels
stub: true # whether to auto-insert stub texts
head: | # header stub text to prepend when indicated
Below the next line is the body...
---
tail: | # footer stub text to append when indicated
---
This issue was automatically generated by issuer.
body: | # body text to impose when no body provided
This is the default text that will appear if an issue record is a _stub_ and no `body` field is designated.
issues: # block for listing issues to post to cloud
- summ: Issue title # title/summary field
type: Task # type of issue (must already be registered)
body: | # description/body field
Markdown-formatted description.
tags: [label1, 'component:api'] # labels to create/assign
user: someusername # assigned username
stub: false
- summ: Another issue
vrsn: 0.2.0 # milestone
# this issue record's body will be:
# Below the next line is the body...
# ---
# This is the default text that will appear if an issue record is a _stub_ and no `body` field is designated.
# ---
# This issue was automatically generated by issuer.
- summ: Documentation issue
tags: [-needs:docs] # skip the default needs:docs label
The $meta
block is entirely optional, but if it is absent, your issuer
command will need a --proj
flag to designate the GitHub repo to which your issues should post.
Only the summ
property is required for each issue record, and issue records (Array items) that are simple strings will be treated as summary-only.
Therefore, the following example would yield 3 tickets with unique summaries and the same body, based on $meta.defaults.body
.
Issuer will prompt the creation of tags (labels) or versions (milestones) if they do not already exist in the target repository.
Any type
entry must correspond to an existing issue type.
$meta:
defaults:
body: |
This is a placeholder text until someone is assigned this ticket.
stub: true
issues:
- summ: Fill out issue bodies and add labels/milestones
body: |
The rest of the tickets in this project are not filled out.
tags: ['priority:high']
- Make a README.adoc file
- Add a license to the repo
💡
|
This repository contains numerous example files to use for inspiration. |
The IMYML format will be standardized and formally specified in a future release of issuer, but it will remain an open standard adoptable by anyone who wants to exploit or extend it.
- $meta
-
Optional block for establishing operation-wide modes and settings.
- $meta.proj
-
(String) Designates the target project/repository.
- $meta.defaults
-
Properties in this block establish the default values to be used for any issue record that does not specify a value for the given property.
- $meta.defaults.vrsn
-
(String) Sets default version or milestone for all issues.
- $meta.defaults.user
-
(String) Sets default assignee (GitHub username).
- $meta.defaults.type
-
(String) Sets default issue type to apply to all issues when no
type
property is specified in the issue record. - $meta.defaults.tags
-
(Array): Labels to append to issues (comma-separated). Prepend items with `+` to indicate they should be appended to existing labels. Items without `+` will only be used for issues with no `tags` designated.
- $meta.defaults.stub
-
(Boolean) Establishes the state whether to insert stub texts (
body
/head
/tail
). - $meta.defaults.body
-
(String) Sets default body text to apply to all issues when no
body
property is specified in the issue record. - $meta.defaults.head
-
(String) Sets default text to insert before the body of all issues for which
stub
. - $meta.defaults.tail
-
(String) Sets default text to insert after the body of all issues for which
stub
.
- issues
-
(Array) Tabular listing of issue records as Array items. If an item is Scalar (not a Map with named keys), the value must be a String and it will be treated as the
summ
(summary/title) property.Otherwise, any
issues
Array items must be Map-formatted “dictionaries” with the following properties:- summ
-
(String, required) A one-line title or summary of the issue.
- body
-
(String) The main body or description text for the issue. Defaults to
$meta.defaults.body
ifstub == true
for the record, in which case, upon submission, will also incorporate any values for$meta.defaults.head
and$meta.defaults.tail
. - vrsn
-
(String) The milestone associated with the issue.
Defaults to
$meta.defaults.vrsn
or elsenull
. - type
-
(String) The type of issue, which must already be registered in the target project or repository. Defaults to
$meta.defaults.type
or elsenull
. - tags
-
(Array of Strings) A listing of specific labels to assign to the issue.
Supports special prefix notation for label management:
-
Regular labels (example:
bug
,priority:high
) are applied based on default tag logic -
Append labels (example:
+urgent
) are always applied to all issues -
Removal labels (example:
-needs:docs
) remove the specified label from the default/appended labels listExample:
tags: [documentation, +critical, -needs:review]
would adddocumentation
andcritical
labels while removing anyneeds:review
label from defaults.
-
- user
-
(String) The system username of the person or bot to which the ticket is assigned.
- stub
-
(Boolean) Accepts
true
orfalse
. Whether to treat the issue as a stub entry, meaning prepend any$meta.defaults.head
text or append any$meta.defaults.tail
text, and in case the ticket has nobody
property, insert the text of$meta.defaults.body
.
issuer [IMYML_FILE] [options]
A source IMYML file is required and can be specified in two ways:
-
Positional argument (most common): Place the file path immediately after
issuer
-
Named option: Use the
--file
option flag to specify the file path
Examples:
issuer my-issues.yml --proj acmeco/widget1 --dry issuer --proj acmeco/widget1 --file my-issues.yml --dry
These options specify the source file and override any corresponding $meta.defaults
properties.
- --file IMYML_FILE
-
IMYML file path (alternative to positional argument).
- --proj ORG/REPO
-
The target project (org/repo or user/repo format for GitHub).
- --vrsn VERSION
-
Sets default milestone for all issues.
- --user USERNAME
-
Sets default assignee (GitHub username).
- --tags TAG[,TAG]
-
Sets labels to apply to issues (comma-separated). Prepend items with `+` to indicate they should be appended to existing labels. Items without `+` will only be used for issues with no `tags` designated.
- --stub [true*|false]
-
Whether to treat all issues as stubs, meaning prepend any
$meta.defaults.head
text or append any$meta.defaults.tail
text, and in case the ticket has nobody
property, insert the text of$meta.defaults.body
.
- --dry, --dry-run
-
Dry-run: print actions but do not post to GitHub.
- --auto-versions, --auto-milestones
-
Automatically create missing milestones/versions without prompting for confirmation.
- --auto-tags, --auto-labels
-
Automatically create missing labels/tags without prompting for confirmation.
- --auto-metadata
-
Automatically create all missing metadata (milestones and labels) without prompting for confirmation. Equivalent to using both
--auto-versions
and--auto-tags
. - --help, -h
-
Prints the usage screen.
- --version
-
Prints the version of
issuer
.
GitHub authentication requires a valid personal access token.
The application will check for environment variables in the following order:
-
ISSUER_API_TOKEN
-
ISSUER_GITHUB_TOKEN
-
GITHUB_ACCESS_TOKEN
-
GITHUB_TOKEN
To create and set a token:
-
In the GitHub Web interface, go to Settings (under your user icon) → Developer Settings (bottom of left menu) → Personal Access Tokens → Fine-grained tokens.
-
Generate a new token with access to All repositories or any Select repositories you wish to post to, and include read/write permissions GitHub Issues (under Repository permissions).
-
Copy the token and set it as an environment variable.
Exampleexport ISSUER_API_TOKEN=github_pat_xxxxxxxxxxxxxxxxxxxxxxxx
Where
github_pat_xxxxxxxxxxxxxxxxxxxxxxxx
is your actual token.
If your GitHub token is stored under any other name, you can alias it inline by prepending to your issuer
command.
For example:
ISSUER_API_TOKEN=$MY_GITHUB_API_KEY issuer my-issues.yml
When using Docker, you can pass any such key into the container this way, using the -e
option: -e ISSUER_API_TOKEN=$MY_GITHUB_API_KEY
.
Issuer automatically logs all API operations for tracking and potential cleanup.
By default, logs are stored in a user-wide directory:
-
Linux/macOS:
~/.config/issuer/logs/
-
With XDG Base Directory:
$XDG_CONFIG_HOME/issuer/logs/
-
Custom location: Set
ISSUER_CONFIG_DIR
environment variable
Example:
# Use custom config directory
export ISSUER_CONFIG_DIR="~/path/to/my/issuer/config"
issuer my-issues.yml
# Logs will be stored in: ~/path/to/my/issuer/config/logs/
Use the management script to view and manage your run logs.
For now, you will need to clone the DocOps/issuer repo in order to work with the run-logs manager.
# List all runs
ruby scripts/manage-runs.rb list
# Show details for a specific run
ruby scripts/manage-runs.rb show run_20250711_143022_abcd
# Clean up all logs (use with caution)
ruby scripts/manage-runs.rb clean-logs
Each run creates a detailed log with:
-
All created issues, milestones, and labels
-
URLs for easy access
-
Run metadata and status
-
Error information if the run failed
📎
|
It is typically safe to delete logs once you are satisfied with your posted issues. Logs are simply kept for easy reversal of mis-postings. |
I developed the 0.1.0 version of this application after trying to use GitHub Copilot to automatically bulk-create issue tickets, which it promises to be able to do but failed me pretty hard at it.
That facility seems like a perfectly inappropriate use of generative AI. It accepted my plan request and pre-drafted ticket content, but then it wanted me to manually add labels and milestones to them, as well as manually click create on each one — even though I had already taken the time to plan and instruct the milestones and labels and the contents were fully prepared.
Additionally, I find myself using different issue-management systems (Jira, GitLab Issues, etc), so I wanted a more platform-agnostic way to handle this problem. With that in mind, I have left the Ruby API and the IMYML model fairly “generic” for extensibility. I will probably adapt the API to other systems in future releases, and I welcome contributions to that effect.
I should note up front that this is the closest I have come to “vibe coding” anything bigger than a local script, let alone shippable production code. Nevertheless, I intervened to make substantial and specific changes at least 100 times before the 0.1.0 release alone, and I rearranged major aspects of the codebase.
I designed the IMYML format and the CLI up front, then I let Claude 4 (via GH Copilot) draft most of the code. It committed lots of rookie mistakes during this process, and it even confessed to “cargo-cult programming” when I pointed out it was introducing some anti-patterns.
In the end, the only thing that is mainly untouched by me are the rspec tests, which I will more fully examine and approve before any 1.0 release, but for now they’ll have to do.
This also explains why the terminal output contains emojis. I will probably make those togglable or configurable in the future.
The specs/
directory contains all specifications, requirements, and tests for the Issuer CLI tool.
specs/
├── tests/
├── rspec/
├── spec_helper.rb
├── cli_spec.rb
├── issue_spec.rb
├── ops_spec.rb
└── issuer_spec.rb
From the project root:
# Run all PR tests locally (same as GitHub Actions)
bundle exec rake pr_test
# Run all tests (recommended)
bundle exec rake spec
# Run all tests directly with RSpec
bundle exec rspec
# Run specific test file
bundle exec rspec specs/tests/rspec/cli_spec.rb
# Run tests matching a pattern
bundle exec rspec --pattern "*ops*"
The pr_test
task runs the exact same tests that GitHub Actions runs for pull requests:
-
RSpec Tests: All unit tests (
bundle exec rake spec
) -
CLI Tests: Command-line interface functionality tests
-
YAML Validation: Validates all example YAML files
-
Documentation Quality: Vale linting on all documentation files
This ensures you can validate your changes locally before pushing to GitHub.
- RSpec Tests
-
(
specs/tests/rspec/
): Executable tests that validate the gem functionality - GitHub API Integration Tests
-
(
specs/tests/github-api/
): Comprehensive end-to-end tests for GitHub API integration - Test Data
-
(
specs/tests/data/
): YAML files, fixtures, and other test assets - Future
-
Natural language specifications, API schemas, requirement documents
For comprehensive GitHub API testing, see the GitHub API test suite documentation.
Quick start for API integration testing:
# Check GitHub connectivity
./specs/tests/check-github-connectivity.sh
# Run full GitHub API integration test suite
./specs/tests/run-github-api-tests.sh
# Run with specific options
./specs/tests/run-github-api-tests.sh --verbose --auto-cleanup
The GitHub API test suite validates:
-
Authentication and connectivity
-
Issue creation with various configurations
-
Milestone/version management and automation
-
Label/tag management and automation
-
Assignment functionality
-
Automation flags (
--auto-metadata
,--auto-versions
, etc.) -
Error handling and edge cases
For detailed API documentation, see the automatically generated documentation at GemDocs.
The API reference includes:
-
Complete class and method documentation
-
Method signatures and parameters
-
Return types and examples
-
Internal implementation details
This documentation is automatically updated with each gem release.
As of version 0.2.0, the Release Notes and Changelog are generated using the ReleaseHx tool, which is still in pre-release.
You can find the release history assets in the docs/releasehx/
directory, which contains a configuration file and the 0.2.0 “RHYML” file that was auto-modified and then manually edited to produce the GitHub release announcement.
ReleaseHx will be available soon, but for now the following is only usable by DocOps Lab.
-
Add local releasehx gem to
Gemfile
:gem 'releasehx', path: '../releasehx'
-
Draft RHYML draft file from the online GitHub Issues for the given version:
./releasehx-install.sh && bundle exec rhx 0.2.1 --yaml docs/releasehx/release-0.2.1.yml --config docs/releasehx/config.yml
The
releasehx-install.sh
ensures ReleaseHx is up to date. -
Manually edit the RHYML draft file.
-
Generate the release history as Markdown.
bundle exec rhx docs/releasehx/release-0.2.1.yml --md docs/releasehx/0.2.1.md --config docs/releasehx/config.yml
-
Copy and paste the contents of
docs/releasehx/0.2.1.md
into the GitHub release form at https://github.com/DocOps/issuer/releases/new.
Bug reports and pull requests are welcome on GitHub at https://github.com/DocOps/issuer.