Structures feature ideas and bug reports into formal PRDs with verifiable acceptance criteria.

Install

mkdir -p .claude/skills/prd-authoring && curl -L -o skill.zip "https://agentskills.codes/api/skills/download/11796" && unzip -o skill.zip -d .claude/skills/prd-authoring && rm skill.zip

Installs to .claude/skills/prd-authoring

Activation

This is the description your AI agent reads to decide when to run this skill — the better it matches your request, the more reliably it fires.

Generate a Product Requirements Document for a feature request, bug report, product change, or other multi-step engineering work. Use when the human asks to plan a feature, write a PRD, define acceptance criteria, or analyze a codebase before implementation. Stops at acceptance-criteria approval — does not implement code.
323 chars✓ has a “when” triggerlonger than Claude Code's old 250-char listing cap (fine on current versions)
Intermediate

Key capabilities

  • Explore the codebase before proposing solutions
  • Generate a written Codebase Analysis artifact
  • Ask clarifying questions with multiple-choice options
  • Draft Product Requirements Documents (PRDs)
  • Present acceptance criteria for human approval
  • Enforce human gates for Codebase Analysis and Acceptance Criteria

How it works

The skill follows a structured process to explore the codebase, generate a codebase analysis, and ask clarifying questions. It then drafts a PRD and presents acceptance criteria for human approval.

Inputs & outputs

You give it
Feature request, bug report, or product brief
You get back
An approved Product Requirements Document (PRD) with verifiable acceptance criteria

When to use prd-authoring

  • Write a PRD for a new feature
  • Draft requirements for a bug report
  • Define acceptance criteria for product changes
  • Analyze codebase before implementation

About this skill

PRD Authoring

Turn a feature request into an approved PRD with verifiable acceptance criteria. The full process is defined in .ai-rules/rules/02-prd.md — this skill is the trigger shape and the gate enforcer. Read that rule first.

When this skill should fire

  • Human asks to "plan a feature", "write a PRD", "spec this out", "draft requirements", or similar.
  • Human pastes a bug report, feature idea, or product brief that needs structured decomposition before implementation begins.
  • An existing PRD needs sharpened acceptance criteria.

Do not fire for: implementation requests, refactors, code review, or work where a human-approved PRD already exists.

Required behavior

Follow .ai-rules/rules/02-prd.md exactly. The short version:

  1. Explore the codebase first. Produce a written Codebase Analysis.
  2. Present the analysis. Wait for the human to confirm it before continuing.
  3. Ask 3–8 clarifying questions with lettered options.
  4. Generate the PRD.
  5. Present acceptance criteria separately. Wait for human approval.
  6. Stop. Do not generate tasks or write implementation code.

Progressive disclosure

Load these references only when you need them:

  • references/codebase-analysis.md — rubric, common failure modes, and a worked example for the Codebase Analysis artifact. Load when producing step 1.
  • references/acceptance-criteria.md — rubric, common failure modes, and a worked example for verifiable acceptance criteria. Load when drafting step 5 or when sharpening vague human-supplied criteria.

The rule file is the source of truth for the process. References here exist for taste and calibration; they do not duplicate the rule body.

Hard gates

These gates are non-negotiable. Stop and prompt the human at each one:

  • Codebase Analysis gate — do not propose a solution before the human confirms your reading of the codebase.
  • Acceptance Criteria gate — do not proceed past the PRD until the human explicitly approves the acceptance criteria.

If the human asks you to skip a gate, refuse and explain why. The whole point of this skill is to preserve the human gates that the conversational flow would otherwise erode.

Out of scope

  • Task generation. After AC approval, hand off to a task-breakdown skill or equivalent in your runner. This skill ends at "AC approved".
  • Implementation. Never edit production code under this skill.

When not to use it

  • For implementation requests
  • For refactors or code review
  • When a human-approved PRD already exists

Limitations

  • Does not generate tasks
  • Does not proceed past the PRD until the human explicitly approves the acceptance criteria

How it compares

This skill enforces a design-first approach with explicit human approval gates for codebase analysis and acceptance criteria, preventing premature implementation.

Compared to similar skills

prd-authoring side by side with the closest alternatives in the catalog.

SkillInstallsUpdatedSafetyDifficulty
prd-authoring (this skill)01moNo flagsIntermediate
planning-with-files2336moReviewIntermediate
trello412moReviewBeginner
pmbok-project-management389moNo flagsIntermediate

Try saying

Example prompts that trigger this skill in your AI assistant.

You might also like

planning-with-files

davila7

Implements Manus-style file-based planning for complex tasks. Creates task_plan.md, findings.md, and progress.md. Use when starting complex multi-step tasks, research projects, or any task requiring >5 tool calls.

233106

trello

openclaw

Manage Trello boards, lists, and cards via the Trello REST API.

41205

pmbok-project-management

jgtolentino

Comprehensive PMP/PMBOK project management methodologies and best practices. Use this skill when users need guidance on project management processes, templates, knowledge areas, process groups, tools, techniques, or certification preparation. Covers all 10 PMBOK Knowledge Areas and 5 Process Groups with practical templates, frameworks, and industry-standard approaches. Includes risk management, stakeholder engagement, schedule management, cost control, quality assurance, and resource planning.

38183

clickup

civitai

Interact with ClickUp tasks and documents - get task details, view comments, create and manage tasks, create and edit docs. Use when working with ClickUp task/doc URLs or IDs.

37176

ma-playbook

alirezarezvani

M&A strategy for acquiring companies or being acquired. Due diligence, valuation, integration, and deal structure. Use when evaluating acquisitions, preparing for acquisition, M&A due diligence, integration planning, or deal negotiation.

39122

create-plan

antinomyhq

Generate detailed implementation plans for complex tasks. Creates comprehensive strategic plans in Markdown format with objectives, step-by-step implementation tasks using checkbox format, verification criteria, risk assessments, and alternative approaches. Use when users need thorough analysis and structured planning before implementation, when breaking down complex features into actionable steps, or when they explicitly ask for a plan, roadmap, or strategy. Strictly planning-focused with no code modifications.

36121

Search skills

Search the agent skills registry