quay/ai-helpers

Perfil de quay

Última atividade de origem registrada: 18 de agosto de 2026 às 15:10

Catálogo do SkillsMP atualizado: 17 de outubro de 2026 às 00:09

skills coletadas: 38

Estrelas no GitHub: 3
Forks no GitHub: 11

README do repositório

Arquivo de origem do README: README.md

Ler o README original

Expandir README completo

# quay-ai-helpers Shared agent toolkit for the Quay organization. Provides reusable Claude Code plugins for JIRA workflow automation, development lifecycle management, and testing infrastructure. ## Plugins ### dev The Ralph Loop: a continuous state machine that takes a JIRA ticket from assignment to merge-ready PR. Includes the full skill chain (`start`, `code`, `pr`, `poll`, `ci`, `backport`) plus the unified `/work` orchestrator. ### jira JIRA operations (view, assign, transition, check/set Target Version) and planning commands for decomposing features into epics, stories, and estimates. Includes safety hooks for embargoed tickets. ### konflux Konflux advisory skills: query curated NotebookLM knowledge bases for Konflux concepts, pipelines, builds, releases, and troubleshooting. Requires the `nlm` CLI (`pipx install nlm`). ### openshift-testing Ephemeral OpenShift cluster provisioning via Gangway API and remote Playwright browser server deployment for E2E testing. ## Installation ### Via Lola (recommended) [Lola](https://github.com/RedHatProductSecurity/lola) is a universal AI package manager that distributes skills across AI assistants. Each plugin in this repo is a Lola-compatible module. #### One-time setup ```bash # Install individual plugins uvx --python 3.13 --from lola-ai lola mod add https://github.com/quay/ai-helpers.git --module-content=plugins/dev uvx --python 3.13 --from lola-ai lola mod add https://github.com/quay/ai-helpers.git --module-content=plugins/jira uvx --python 3.13 --from lola-ai lola mod add https://github.com/quay/ai-helpers.git --module-content=plugins/konflux uvx --python 3.13 --from lola-ai lola mod add https://github.com/quay/ai-helpers.git --module-content=plugins/openshift-testing # Install to your project lola install dev -a claude-code ./my-project lola install jira -a claude-code ./my-project ``` #### Declarative dependencies Add a `.lola-req` file to your project root: ``` # .lola-req — AI context modules for this project https://github.com/quay/ai-helpers.git@main --module-content=plugins/dev https://github.com/quay/ai-helpers.git@main --module-content=plugins/jira https://github.com/quay/ai-helpers.git@main --module-content=plugins/konflux https://github.com/quay/ai-helpers.git@main --module-content=plugins/openshift-testing ``` Then sync all modules: ```bash uvx --python 3.13 --from lola-ai lola sync ``` #### What Lola installs | Asset type | Destination | |-----------|-------------| | Skills (SKILL.md) | `.claude/skills/{name}/SKILL.md` | | Scripts (*.sh) | `.claude/scripts/` (via post-install hook) | | Templates | `.claude/templates/` (via post-install hook) | | Commands (*.md) | `.claude/commands/` (via post-install hook) | #### Updating ```bash lola mod update dev # Pull latest from source lola install dev -a claude-code --force # Reinstall ``` ### Via Claude Plugin ```bash claude plugin marketplace add quay/ai-helpers ``` ## Configuration All project-specific values are set via environment variables with Quay defaults. See each plugin's README for the full variable list. ### Core Variables | Variable | Default | Purpose | |----------|---------|---------| | `JIRA_DOMAIN` | `redhat.atlassian.net` | JIRA instance | | `PRIMARY_BRANCH` | `master` | Main branch name | | `DEFAULT_REPO` | `quay/quay` | GitHub org/repo | | `PR_TITLE_PATTERN` | PROJQUAY/QUAYIO regex | CI-enforced PR title regex | | `JIRA_TARGET_VERSION_FIELD` | `customfield_10855` | Target Version field ID | ### Hook Setup Copy `plugins/dev/templates/settings.json.template` to your project's `.claude/settings.json` and adjust script paths to reference the plugin install location. ## Project Structure ``` ai-helpers/ ├── plugins/ │ ├── dev/ # Ralph Loop + dev lifecycle │ │ ├── skills/ # start, code, pr, poll, ci, backport, work │ │ ├── scripts/ # Shell scripts for hooks and automation │ │ ├── templates/ # PR description, settings.json template │ │ └── lola.yaml # Lola module metadata │ ├── jira/ # JIRA ops + planning commands │ │ ├── skills/ # jira │ │ ├── scripts/ # jira-ops, embargo checks, etc. │ │ ├── commands/ # 8 planning commands │ │ └── lola.yaml # Lola module metadata │ ├── konflux/ # Konflux advisory (NotebookLM) │ │ ├── skills/ # konflux-advisor │ │ ├── scripts/ # Lola post-install hook │ │ └── lola.yaml # Lola module metadata │ └── openshift-testing/ # Cluster + browser testing │ ├── skills/ # cluster-provision, remote-playwright │ ├── scripts/ # Provisioning scripts │ └── lola.yaml # Lola module metadata ├── scripts/ │ └── lola-post-install.sh # Shared post-install hook (symlinked by plugins) ├── templates/ # Starter files for adopting repos │ ├── AGENTS.md.template │ └── CLAUDE.md.template ├── docs/ # Marketplace documentation site └── Makefile ``` ## Adoption Guide 1. Install the plugin: `claude plugin add quay/ai-helpers` 2. Set project-specific env vars in your `.claude/settings.json` or shell profile 3. Copy `templates/AGENTS.md.template` to create your project's `AGENTS.md` 4. Copy `plugins/dev/templates/settings.json.template` for hook configuration 5. Use `/dev:work PROJQUAY-XXXX` to run the full development lifecycle ## Development ```bash make lint # Validate plugin structure make update # Regenerate docs make new-plugin NAME=foo # Create a new plugin ``` ## License MIT

Carregando o README do repositório…

Skills neste repositório

Desenvolvedores de software Analistas de garantia de qualidade de software e testadores Administradores de redes e sistemas de computador Especialistas em gestão de projetos

6 categorias ocupacionais · 100% classificado

skill

ocupação

descrição

atualizado

quay-retro-signal-filter

Analistas de garantia de qualidade de software e testadores

Use with the retro agent when deciding whether workflow observations are actionable proposals. Filters process-only noise and deduplicates proposals by root cause while preserving concrete production, security, CI, and recurring contributor failures.