agentleFS
Sign inStart free
fapariciorfaparicior/sdlc-skills/skills/discovery/prd-workshop/SKILL.md

prd-workshop

Skill2 starsChanged 7 days agoMITView on GitHub

Draft and validate formal Product Requirements Documents (PRDs). USE THIS SKILL when user mentions: prd, product requirements document, create prd, draft prd, write prd, validate prd, review prd, or critique prd. Supports PRD generation from requirements and critical quality validation.

The file

We split this skill into the rules it gives an agent

2 rules. About 0 are common, 1 is distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Ensure all 10 numbered sections from templates/template-prd.md are populated in exact order:

    line 41
  2. Distinctive:

    Before ending your turn, verify every item below with concrete evidence (not just a checkmark).

    line 84

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 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

PROTOCOL: PRD-WORKSHOP (EXECUTION LEVEL 3)

<state_machine> CURRENT_STEP -> REQUIRED_TOOL -> REQUIRED_OUTPUT -> MANDATORY_HALT Workflow A (Draft) -> read_file(templates/template-prd.md) -> Complete PRD conforming to template-prd.md -> 🛑 HALT TURN (Present PRD) Workflow B (Validate) -> read_file(references/prd-validator.md) -> 5-dimension scored critique (1-10) + blockers -> 🛑 HALT TURN (Present critique) </state_machine>

<execution_invariants>

  1. WORKFLOW IMMUTABILITY: You may reason freely INSIDE each section, but you are strictly forbidden from skipping sections or inventing a custom PRD or validation format (workflow structure = immutable | step reasoning = flexible).
  2. TEMPLATE LOCK:
    • Workflow A (Draft PRD): You MUST call read_file on templates/template-prd.md before writing the PRD, preserve every section heading and table schema verbatim, and strip all instructional meta-commentary.
    • Workflow B (Review PRD): You MUST call read_file on references/prd-validator.md before scoring and follow its exact 5-dimension evaluation rubric.
  3. ESCAPE HATCH: If the situation does not fit the contract, do not silently improvise and do not force-fit: declare the deviation in one line and ask. </execution_invariants>

<output_contract>

  • DRAFT PRD SOURCE_TEMPLATE: templates/template-prd.md (MUST call read_file before writing)
  • REVIEW PRD SOURCE_RUBRIC: references/prd-validator.md (MUST call read_file before scoring)
  • SCHEMA_INVARIANCE: Preserve all 10 numbered template headings and tables verbatim; strip all instructional meta-commentary.
  • ESCAPE_HATCH: Never silently omit or invent PRD sections. If one of the 10 sections genuinely does not apply, preserve the numbered heading and write "N/A — [1-line rationale]", or halt and ask the user before deviating. </output_contract>

1. Supported Workflows

Workflow A: Draft a PRD

  • Generate a structured PRD using templates/template-prd.md.
  • Ensure all 10 numbered sections from templates/template-prd.md are populated in exact order:
    1. ## 1. Executive Summary
    2. ## 2. Problem Statement (2.1 User Pain Points, 2.2 Strategic Alignment / Business Impact)
    3. ## 3. Goals & Success Metrics (3.1 Success Metrics table with Baseline/Target, 3.2 Anti-Goals)
    4. ## 4. User Personas & Stories (4.1 Target Personas, 4.2 Key User Stories & Acceptance Criteria)
    5. ## 5. Functional Requirements (5.1 Core Logic, 5.2 Edge Cases & Error Handling)
    6. ## 6. UX/UI & Design Requirements
    7. ## 7. Technical Considerations (Engineering Specs)
    8. ## 8. Non-Functional Requirements (NFRs)
    9. ## 9. Go-to-Market & Release Strategy
    10. ## 10. Open Questions & Risks
  • Strip all instructional commentary from the output.

Workflow B: Review & Validate a PRD

  • Critically audit a submitted PRD using references/prd-validator.md.
  • Evaluate against 5 dimensions: Clarity, Completeness, Feasibility, Measurability, and Edge Cases.
  • Deliver an actionable critique with a readiness score (1–10), critical blockers, ambiguity quotes, missing edge cases, and clarifying questions.

2. Natural Language Activation

IntentNatural Language ExamplesAction Executed
Draft PRD"Create a PRD for [feature]", "Write a product requirements document", "Draft PRD from inception"Generates a new PRD using templates/template-prd.md.
Review PRD"Review this PRD", "Validate my PRD", "Critique the product requirements document"Audits the PRD against references/prd-validator.md and provides scored feedback.

3. Review Criteria (Quality Checklist)

CategoryStandard
ClarityIdentify vague, non-quantifiable language (e.g., "fast," "easy," "seamless").
CompletenessEnsure all core sections are present and detailed.
FeasibilityFlag requirements that are technically contradictory or overly complex.
MeasurabilityEnsure success metrics are quantifiable and tied to the problem statement.
Edge CasesCheck that error states, empty states, and offline modes are documented.

<self_validation_gate> Before ending your turn, verify every item below with concrete evidence (not just a checkmark). If any item cannot be cited with real evidence from this turn, DO NOT produce your final response — fix the violation first:

  • Template/Rubric Read Evidence: Exact path passed to read_file (templates/template-prd.md for Workflow A or references/prd-validator.md for Workflow B): <path>
  • Schema & Escape-Hatch Evidence: Confirm all 10 numbered PRD sections (1–10) or all 5 validation rubric dimensions are present verbatim (noting any "N/A — [rationale]" section): <10/10 or 5/5 status> </self_validation_gate>

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.

Check your setup, free