Create & distribute PF AI tooling to GE workflows

[patternfly-jira-sync]

Problem

Multiple teams (Firefly, UXD SDLC, others) are independently building AI skills, agents, and plugins for software delivery workflows. No shared distribution mechanism exists. Teams duplicate work, patterns diverge, and there's no path from "skill in a repo" to "tool in an engineer's workflow."

Goal

1. Create skills — Build and publish reviewed skills into patternfly/ai-helpers, uxd/ai-helpers, or other appropriate locations. This is the primary deliverable.
2. Establish governance & distribution — Define ownership boundaries, contribution process, and a distribution model that lets other workflows (Ansible, Firefly, etc.) pull in skills we create and maintain.

Success = skills exist, are published, and other teams can demonstrably consume them in their own pipelines.

Scope

Spikes investigate patterns, architecture, and pruning. Follow-up issues deliver skills and distribution infrastructure. Implementation epics follow from spike outcomes.

Key questions this epic answers

3. Do multi-agent subagent patterns justify their overhead vs single-agent skills?
4. How do downstream teams consume ai-helpers without duplicating or forking?
5. Which of the various proposed skills actually deserve prioritization?
6. How can we refine skills so they perform efficiently and predictably?
7. What's the architecture boundary between patternfly/ai-helpers and uxd/ai-helpers, and how do both feed a marketplace?

---

Jira Issue: PF-4038

[coderabbitai]

🔗 Related PRs

#26 - feat(skill): summarize jira issues & PR reviews [merged]
#48 - Add Jira quarterly-initiative-report skill to pf-workflow plugin [merged]
#53 - chore(skills): Update duplicate-epic skill to work with bugs and stories [merged]

---

📝 Issue Planner

_{Check the box below or use the @coderabbitai plan command to generate an implementation plan and prompts that you can use with your favorite coding assistant.}

• Create Plan

---

🧪 Issue enrichment is currently in open beta.

You can configure auto-planning by selecting labels in the issue_enrichment configuration.

To disable automatic issue enrichment, add the following to your .coderabbit.yaml:

issue_enrichment:
  auto_enrich:
    enabled: false

💬 Have feedback or questions? Drop into our discord!