Engineering PRD: Scope to Acceptance Criteria

    1

    Turn a rough feature idea into clear requirements, scope boundaries, and testable acceptance criteria.

    Free

    2 usesSecurity scanned

    Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Engineering PRD: Scope to Acceptance Criteria

    Example session with this skill installed

    Write a short PRD for archiving completed tasks in our freelancer task app. Only the task owner can archive or restore. No bulk actions or permanent deletion.

    • Read your context and instructions
    • Compiled the engineering prd

    The brief below defines owner-only archiving and restoration, with one storage assumption to confirm.

    PRD: Archive Completed Tasks

    Problem and outcome

    Freelancers need to clear finished work from their task list without losing it. Success means completed tasks can be hidden and restored without changing their contents.

    Scope and assumption

    Include single-task archive and restore. Exclude bulk actions and permanent deletion.
    Assumption: Tasks already store an owner and completion status.

    Workflow and requirements

    • Owners see Archive on completed, unarchived tasks.
    • Archiving removes the task from the default list and adds it to an Archived view.
    • Restore returns the task to the default list, still completed.
    • Store an archive timestamp; clear it on restore. Preserve all other task data.
    • Enforce ownership on the server, not just through hidden buttons.

    Acceptance criteria

    • An owner archives a completed task and finds it in Archived after reload.
    • Restoring preserves its title, details, and completed status.
    • Archiving an incomplete task is rejected without changes.
    • A non-owner cannot archive or restore through a direct request.
    • Failed saves show an error and leave the task in its prior state.

    Risk and implementation outline

    List filters could hide tasks incorrectly. Add archive state, update list queries, add controls, then test both views and permission checks.

    Next steps

    • Confirm the storage assumption before implementation, a blocking decision.
    • Review the brief, then break the approved scope into tasks.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    A rough feature idea leaves too much room for guesswork. Turn it into a product requirements document (PRD) that explains what to build, what to leave out, and how to check that it works.

    What it does

    • Clarifies the request: Identifies the users, their problem, and the intended result.
    • Defines the scope: Separates essential work from extras and future ideas.
    • Makes requirements testable: Describes expected behavior and acceptance criteria your team can verify.
    • Flags missing decisions: Calls out assumptions, risks, edge cases, and unanswered questions.
    • Supports handoff: Adds an implementation outline without jumping straight into code.

    How it works

    1. Share your idea, feature request, design discussion, or rough plan.
    2. Answer focused questions if key details are missing. Otherwise, receive a draft with assumptions clearly marked.
    3. Review the requirements, scope, and open decisions with your team.
    4. Use the document to guide implementation or a separate task breakdown.

    Works with

    Plain-language requests and existing product or engineering notes. The resulting document can be reviewed by people and used as input for Codex or an engineering team.

    Why use this skill

    Spend less time explaining the same feature across messages. Give reviewers concrete behavior to approve and testers clear conditions to check before work is considered complete.

    Use cases

    • Prepare a developer brief for a new app feature.
    • Turn a design discussion into agreed requirements.
    • Define a smaller first release for an oversized request.

    Known limitations

    Missing details may require clarification. This skill focuses on requirements and an implementation outline, not a full task tracker or automatic implementation.

    Source and license

    Created by Ray Carter (ruicore). Free and open source under Apache-2.0.

    Agensi serves the latest approved marketplace version, which may differ from GitHub main. GitHub changes do not automatically update the marketplace package.

    Connect via https://www.agensi.io/mcp and ask for this skill by its published title.

    Missing details may need clarification. Produces requirements and an implementation outline, not a full task tracker; code is outside the default scope.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before any payment.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    2 uses

    Times used to date

    Free forever

    No account required to browse

    Trust & safety

    Security scan

    Security scan passed 23 days ago

    • Free with an Agensi account

    Permissions

    No special permissions declared or detected

    Listed1 month ago
    Updated23 days ago

    What's inside

    Frequently Asked Questions

    This skill is designed to take unstructured feature ideas or messy meeting notes and transform them into a structured PRD with clear scope, defined user roles, and measurable acceptance criteria.

    Designed for Codex. Other agents that can load SKILL.md may follow the workflow, with output shaped by their own tools and context.

    The skill is optimized to produce high-quality drafts from minimal input; however, if your request is missing critical context like target users or specific constraints, the agent will prompt you with focused questions to fill those gaps.

    A SKILL.md workflow for turning rough ideas into scoped, testable PRDs. The public GitHub source is linked above.

    Yes, one of the core functions of this skill is to define 'out-of-scope' items, helping you strip away non-essential features to create a lean, testable first release.

    The GitHub repository is the public source. Marketplace updates are available only after a new version is submitted and approved.

    New in Technical Documentation