agentleFS
PhlegonlabsPhlegonlabs/product-delivery-harness/skills/product-definition-builder/SKILL.md

product-definition-builder

Skill1 starsChanged 18 days agoMITView on GitHub

Use only when the user explicitly requests product-definition-builder or a requested skill workflow requires it. Create or refine owner-approved Product Definition packages from user needs, including discovery, measurable requirements, UI surface behavior, data/trust and AI gates, implementation-ready backend and frontend architecture, coherent owner-approved technology and commercial stack decisions, release targets, tests with executable scenarios, and sourced market reconciliation. Use for PRDs, product specs, architecture, backend/data/auth/deployment choices, frontend stack choices, and product research. It stops at Product Definition Approval and hands UI-bearing products to ui-design-builder; it does not create wireframes or visual design.

The file

We split this skill into the rules it gives an agent

7 rules. About 0 are common, 7 are distinctive, and 0 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈7

    Distinctive · estimate

    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:

    Run only on the user's explicit request, continuation, or a required dependency of that requested skill workflow.

    line 8
  2. Distinctive:

    Read the owning section before its matching action.

    line 28
  3. Distinctive:

    Before a definition or enhancement starts, read Workflow and Interview Rules.

    line 30
  4. Distinctive:

    Before Product Definition Approval is requested, read Self-Review Before Product Definition Approval and run its independent read-only PRD …

    line 31
  5. Distinctive:

    Before a UI handoff is prepared, read Current UI Preflight.

    line 32
  6. Distinctive:

    Before applicable domain decisions are resolved, read the guides that Workflow step 9 names.

    line 33
  7. Distinctive:

    No UI design, implementation, PLAN/RUN, or E2E evidence register starts here.

    line 43

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “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 Definition Builder

Run only on the user's explicit request, continuation, or a required dependency of that requested skill workflow. A matching task, installation, discussion, source edit or existing project artifact does not activate it. Required checks and dependencies stay within the requested outcome; a new top-level stage needs an explicit request.

Installed Commands

Resolve <product-definition-builder-skill-root> and logical script paths as Installed Commands states.

Overview

At every invocation, run the entry checklist.

For a UI enhancement, hand off the added/changed UI IDs and necessary entry or return paths, with all other surfaces explicitly preserved. Follow ../ui-design-builder/references/enhancement-recommendations.md's Incremental UI Scope for the downstream boundary; do not request a fresh all-screen design package or reset unaffected approvals.

Before approval, define test personas (role, tenant, account/data state), applicable Web/native/agent journeys, observable success and denial/no-side-effect assertions, and mock versus real-sandbox obligations in the TEST rows, their scenarios in docs/product/features/, and the architecture testing section. Record non-applicability; never invent login for a headless or unauthenticated product. The downstream Harness freezes the executable matrix and owns fixtures and evidence through ../delivery-harness/references/delivery-acceptance-contract.md; this skill creates no accounts or evidence register.

Use this skill to turn a product idea into a decision-ready package, not an agent-selected implementation guess. Define product scope, journeys, UI surface responsibilities, routes, actions, states, data, trust, and security boundaries, acceptance signals, implementation architecture, release targets, and technology constraints. Draft and reconcile PRD.md with its feature files, architecture.md, and stack-decisions.md, checking scenario coverage with scripts/check_feature_coverage.py while drafting; present coherent frontend, backend/data/auth, mobile/desktop, AI, deployment, and commercial technology options; then obtain an explicit Stack Decision Checkpoint and Product Definition Approval. Stop there. A later explicit UI request invokes ui-design-builder; a later explicit delivery-harness request implements approved inputs.

Stage Routing

Read the owning section before its matching action. Domain guides stay conditional.

  • Before a definition or enhancement starts, read Workflow and Interview Rules. Enhancement mode first uses its scoped delta branch and does not force an initial interview.
  • Before Product Definition Approval is requested, read Self-Review Before Product Definition Approval and run its independent read-only PRD review.
  • Before a UI handoff is prepared, read Current UI Preflight.
  • Before applicable domain decisions are resolved, read the guides that Workflow step 9 names.
  • Actual initial drafting follows the complete required procedure in Workflow.

Entry Boundary

  • PRD.md and architecture.md remain the canonical English implementation sources; complete Chinese review copies remain required.
  • Product Definition Approval and the Stack Decision Checkpoint are explicit owner gates.
  • The PRD ## At a Glance records the owner-selected Delivery profile: standard | mvp | program. Under mvp, the Stack Decision Checkpoint and Product Definition Approval are one owner call; both decisions stay explicit.
  • A program package is a program PRD, architecture and shared stack plus one package per module (Program Packages). It validates per module and once at program level. Module delivery runs one Epic and one PLAN/RUN per module, in Module Map order.
  • Enhancement preserves the same current PRD and stable IDs.
  • No UI design, implementation, PLAN/RUN, or E2E evidence register starts here.
  • A completed package does not launch downstream work automatically. The package creates no wireframe, design system, production component, or Harness mission map.

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.