AI specialists for Product Owners and Product Managers. Eight purpose-built agents that work with your Jira to handle the repetitive work — backlog hygiene, triage, delivery tracking, roadmap analysis, market problem definition, and stakeholder communication — so you can focus on decisions.
Every agent learns your team's specific Jira setup (project key, statuses, fields, conventions) through a one-time interview. No hardcoded assumptions.
Clone the repo and point your AI assistant at it:
git clone https://github.com/bmbouter/redhat-agents.gitClaude Code — copy the skills and agents into your project:
cp -r redhat-agents/module/skills/ .claude/skills/
cp -r redhat-agents/module/agents/ .claude/agents/
cp -r redhat-agents/module/commands/ .claude/commands/Cursor — same idea, different target:
cp -r redhat-agents/module/skills/ .cursor/skills/
cp -r redhat-agents/module/agents/ .cursor/agents/
cp -r redhat-agents/module/commands/ .cursor/commands/Or just open the repo directly in your AI tool and reference the skills and agents from there.
Install with Lola (optional)
If you use Lola, it handles the file placement automatically for any supported assistant:
lola mod add https://github.com/bmbouter/redhat-agents.git
lola install redhat-agents -a claude-code # or cursor, gemini-cli, opencodeOr register the marketplace for easier discovery:
lola market add rh-agents https://raw.githubusercontent.com/bmbouter/redhat-agents/main/marketplace.yml
lola install redhat-agents -a claude-codeOr add to your project's .lola-req for declarative team setup:
https://github.com/bmbouter/redhat-agents.git@main
After installation, run two setup commands:
/jira-setup # Configure Atlassian API token + MCP tools
/process-interview # Teach agents your Jira workflow
The process interview generates local/jira-workflow.md — a local, gitignored config that all agents read.
Tactical, Jira-focused agents for day-to-day backlog operations.
@process-discoverer — Workflow Discovery
Interviews you about your Jira setup and generates the local workflow config. Run this first.
Skills:
process-interview
@backlog-curator — Backlog Health
Scans for stale tickets, missing fields, duplicates, and wrong statuses. Prepares grooming agendas. Suggests priority rebalancing when priorities drift.
Skills:
backlog-health-check·grooming-prep·priority-rebalance
@intake-triager — Incoming Work
Classifies new tickets (bug vs task vs epic), searches for duplicates, assesses severity and effort, drafts acceptance criteria. Recommends priority and parent epic.
Skills:
intake-triage·impact-assessment·acceptance-criteria-writer
@delivery-tracker — Progress & Risks
Tracks epic completion, flags at-risk work, maps dependency chains, generates stakeholder-ready status reports. Read-only — reports and recommends, never modifies tickets.
Skills:
epic-health·status-report·dependency-mapper
Strategic, outward-facing agents for roadmap, market analysis, and stakeholder communication.
@roadmap-strategist — Roadmap Planning
Reviews roadmap health (overcommitment, stalled epics, timeline conflicts). Helps plan upcoming quarters. Generates roadmap briefs tailored to executives, engineering, or customers.
Skills:
roadmap-review·quarter-planning·roadmap-brief
@requirements-analyst — Specs & Signals
Turns unstructured input (conversations, feature requests, support escalations) into structured feature specs. Aggregates customer signals from Jira to surface demand patterns.
Skills:
feature-spec-writer·customer-signal-aggregator
@market-analyst — Market Intelligence
Structures market problems (segments, pain points, alternatives, opportunity sizing). Maps competitive landscapes. Scores feature opportunities against demand, competition, strategic fit, and feasibility.
Skills:
market-problem-definition·competitive-landscape·opportunity-assessment
@stakeholder-communicator — Outward Communication
Documents product decisions with full context and alternatives. Drafts customer-facing release notes from Jira data. Prepares executive briefs adapted to the audience.
Skills:
decision-record·release-notes-drafter·stakeholder-brief
| Skill | What it does |
|---|---|
| jira-setup | One-time Atlassian API token + MCP configuration |
| process-interview | Generates local/jira-workflow.md from an interactive interview |
| suggest-improvement | Files issues/PRs against this repo when agents find gaps |
| backlog-health-check | Scans for stale, incomplete, orphaned, or duplicate tickets |
| grooming-prep | Prepares grooming session agendas with readiness assessments |
| priority-rebalance | Reviews priority distribution and suggests adjustments |
| intake-triage | Classifies tickets, finds duplicates, recommends fields |
| impact-assessment | Evaluates severity, urgency, and effort with Jira context |
| acceptance-criteria-writer | Drafts testable acceptance criteria from descriptions |
| epic-health | Reports epic completion, velocity, at-risk issues, target dates |
| status-report | Generates stakeholder status reports from Jira data |
| dependency-mapper | Traces blocking chains and ranks high-impact blockers |
| roadmap-review | Analyzes roadmap for overcommitment, gaps, and stalled epics |
| quarter-planning | Plans upcoming quarters with scope estimates and dependencies |
| roadmap-brief | Translates epic data into audience-tailored roadmap narratives |
| feature-spec-writer | Converts unstructured input into structured feature specs |
| customer-signal-aggregator | Mines Jira for demand patterns grouped by theme |
| market-problem-definition | Structures market problems with segments, pain points, sizing |
| competitive-landscape | Maps competitors by capability area with gap analysis |
| opportunity-assessment | Scores feature opportunities across 5 dimensions |
| decision-record | Documents decisions with context, alternatives, and rationale |
| release-notes-drafter | Generates customer-facing release notes from closed Jira work |
| stakeholder-brief | Prepares executive summaries adapted to the audience |
Every agent includes the suggest-improvement skill. When an agent discovers a gap, error, or missing capability in its own workflows, it files a GitHub issue (and optionally a PR) against this repository. The user is always informed and can veto before filing.
This creates a feedback loop: real-world usage continuously improves the agents.
module/
├── AGENTS.md # Module overview for AI assistants
├── mcps.json # MCP server configuration
├── skills/ # 23 skills (each in <name>/SKILL.md)
├── agents/ # 8 agent definitions
└── commands/ # Slash commands (jira-setup, process-interview)
marketplace.yml # Lola marketplace definition (optional)
.lola-req # Declarative Lola installation (optional)
local/ # Your Jira workflow config (gitignored)
Apache License 2.0 — see LICENSE for details.