← Back to Payloads
Engineering2026-04-23

agile-po: The ARCHITECT AI That Owns Your Sprint Backlog

Meet agile-po — an ARCHITECT-tier Claude Code blueprint that acts as your AI Product Owner. It writes user stories, prioritizes backlogs with WSJF scoring, runs sprint planning, and generates burndown charts from Git commit data.
Quick Access
Install command
$ mrt install agile
agile-po: The ARCHITECT AI That Owns Your Sprint Backlog
**TL;DR:** `agile-po` is an ARCHITECT-tier Claude Code blueprint that functions as an AI Product Owner — writing user stories, scoring backlog priority with WSJF, running sprint planning sessions, and tracking velocity from Git history. Fully autonomous sprint management in under 3 minutes.

The 10-Second Pitch

  • **WSJF backlog scoring** — automatically ranks features by cost-of-delay / job-size
  • **AI user story generation** — converts feature descriptions into INVEST-compliant stories with acceptance criteria
  • **Sprint planning automation** — takes a prioritized backlog and outputs sprint goals, capacity-adjusted assignments, and burndown forecasts
  • **Git-powered velocity tracking** — pulls commit data to calculate team velocity, forecast delivery dates, and flag at-risk sprints
  • **ARCHITECT tier** — orchestrates multiple Agile tools and frameworks simultaneously, not just one

Setup Directions

Step 1 — Install the Blueprint

```bash

mrt install "agile-po"

```

Step 2 — Initialize Your Project Context

Create `agile-po.config.json` in your project root:

```json

{

"team_velocity": 42,

"sprint_weeks": 2,

"team_size": 6,

"focus_factor": 0.75,

"wsjf_mode": "auto"

}

```

Step 3 — Generate Your Sprint Backlog

```bash

claude -- blueprint agile-po --goal "Reduce checkout abandonment" --sprint 3

```

Step 4 — Review Generated Sprint Plan

Output includes:

  • **User stories** with story points and acceptance criteria
  • **WSJF priority ranking** with cost-of-delay scores
  • **Sprint capacity forecast** adjusted for PTO, meetings, focus factor
  • **Burndown chart data** (Markdown table for import into Jira/Linear)

The Exact Prompt for Mid-Sprint Rescue

```

Our sprint is at risk. Two stories blocked, velocity dropped 30%.

Analyze the current backlog, re-prioritize using WSJF, and generate

a revised sprint plan that salvages the sprint goal. Include revised

burndown and specific unblock actions.

```

Pros & Cons

| Pros | Cons |

|------|------|

| WSJF scoring eliminates gut-feel prioritization | Requires accurate input velocity data |

| User stories written in INVEST format immediately | Story point estimation still needs team calibration |

| Git-powered velocity is always current, not estimated | Doesn't replace Scrum Master facilitation |

| Outputs directly importable to Jira, Linear, Asana | May over-rotate on velocity metrics vs. outcome metrics |

| ARCHITECT orchestration handles sprint retrospective too | Focus factor calibration takes 2-3 sprints to stabilize |

Verdict

`agile-po` earns its ARCHITECT badge — it doesn't just write stories, it runs the whole Agile ceremony stack. The WSJF scoring alone saves hours of backlog grooming, and Git-powered velocity tracking means the sprint forecast is always based on real data, not estimates. If your team is doing Agile theater (ceremonies for their own sake), this blueprint makes them actually functional.

Best for: Engineering managers running Scrum, Product Owners managing large backlogs, and CTOs who want data-driven sprint forecasts without manual chart maintenance.

Alternative: For lightweight story writing without WSJF, use the `content-creator` blueprint. For stakeholder-facing sprint demos, pair with `pr-reviewer` to auto-generate sprint review slides from commit history.

ARCHITECT-tier blueprint. Available at [mr.technology/blueprints](/blueprints). Pro access required.