agentleFS
bookchaowalitbookchaowalit/platform-agent-skills/product-owner/SKILL.md

product-owner

Skill0 starsChanged 9 days agoApache-2.0View on GitHub

Own the Product Owner (PO) role for Solo Empire products — the ordered backlog in tasks-api, the written ordering rule, refinement until items meet the Definition of Ready, iteration or increment goals, accept/reject decisions on delivered items with evidence, and intake of stakeholder feedback into backlog items. Use when ordering or grooming a backlog, deciding what goes into the next iteration, checking DoR/DoD, or accepting a story, demo, or increment. Not for PRDs, roadmaps or KPIs (product-management), detailed requirements and RTM (business-analysis), or schedule, RAID and status reporting (project-management).

12 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

2 rules. 0 are common, 2 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 12 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the ordering rule before ordering: method (WSJF, value ÷ effort, MoSCoW within phase, or risk-first), inputs and where each comes …

    line 18
  2. Distinctive:

    Never accept an item without evidence against its acceptance criteria.

    line 50

“Common” is measured: kept by at least 6 in 10 of the 12 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Product Owner

Turn the approved product direction into an ordered, ready backlog, and decide whether each delivered item is accepted.

Workflow

  1. Establish context: the product's PRD and roadmap version (product-management), the KPI plan, the DoR/DoD in use, and the live backlog in tasks-api (the single backlog, owner decision D2). Open or create the record from architecture-docs templates/positions/product-owner/backlog-and-acceptance-record.md.
  2. Write the ordering rule before ordering: method (WSJF, value ÷ effort, MoSCoW within phase, or risk-first), inputs and where each comes from (KPI ID, user evidence, deadline, dependency, risk), tie-break, and which items bypass ordering (incidents, security, legal).
  3. Order the top of the backlog by that rule. Every item in the top 10 traces to a requirement (FR/US in the BA RTM) or a KPI ID; an item that traces to neither is a discovery question, not a ranked item.
  4. Refine: split items until each fits one iteration, make acceptance criteria testable with business-analysis (user-story.md), get an estimate (ai-native-estimation for multi-module work), and check templates/sdlc/definition-of-ready-and-done.md. Record the change and any open question in the refinement log.
  5. Set the iteration or increment goal as one outcome sentence, commit only items that meet the DoR, and note the capacity basis. An item that misses a DoR condition enters only with an exception approved under the DoR exception policy (owner approval for spec exceptions); record the approver, date, and reference. A recorded gap alone does not admit it.
  6. Accept: compare each delivered item with its acceptance criteria and the DoD, using evidence (tests from qa-testing, a demo, UAT, screenshots, a flow-video recording). Record accept, reject, or accept-with-defect, with the reason. A rejected item or a found defect becomes a backlog item with its ID in the record.
  7. Feedback intake: put every piece of user, client, or review feedback into the intake table with a disposition (new item, merge, decline) and an item ID, and record whether the source was told.
  8. Scope change: when acceptance or feedback changes an approved baseline, raise a change request through business-analysis; do not edit the baseline in place.

Decision rules

  • Order by evidence and the written rule, not by who asked last.
  • Never accept an item without evidence against its acceptance criteria. "Looks done" or "tests pass" alone is not acceptance.
  • Product direction, pricing, and client commitments belong to the product manager or owner; the PO records them as questions until decided.
  • For autopilot repos, new items go in as autopilot:proposed; only the owner moves them to autopilot:todo.
  • For client work, the client's named acceptor signs acceptance; the PO prepares the evidence and records the decision.

Handoff

Report the ordering rule version, the top-of-backlog snapshot with trace links, items refined and still not ready (with the gap), the iteration goal and whether it was met, acceptance decisions with evidence links, new backlog IDs from rejections and feedback, open change requests, and who must decide what next.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.