Agent skill
prd-schema
Use when validating PRD structure, scoring PRD quality, or ensuring product requirements meet the 100-point framework.
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/other/prd-schema
SKILL.md
PRD Schema
Defines the required structure for PRDs created by /arc:specialized:prd. PRDs extend Arc's plan-schema with product-specific sections.
PRD Structure
Required Sections
- Executive Summary — Vision statement and key value proposition (2-4 sentences)
- Problem Statement — Quantified by user segment, with current impact metrics
- Goals & Metrics — SMART goals organized by P0/P1/P2 priority with success metrics table
- Non-Goals — Explicit boundaries of what is NOT included
- User Personas — 2-3 evidence-based personas with goals, frustrations, behaviors
- Functional Requirements — Numbered FR-001 format, priority-ordered, with acceptance criteria
- Implementation Phases — Dependency-ordered phases mapping to Arc plan structure
- Risks & Mitigations — Risk matrix with likelihood, impact, and mitigation strategy
Optional Sections
- Technical Architecture — High-level system design if relevant
- Data Requirements — Data models, storage, privacy considerations
- Integration Points — External systems and API dependencies
- Accessibility Requirements — WCAG compliance targets
- Internationalization — Language and locale requirements
Scoring Framework (100 points)
AI-Specific Optimization (25 pts)
- Structured for AI consumption (clear headings, numbered requirements)
- Machine-parseable acceptance criteria (Given/When/Then or equivalent)
- Unambiguous language (no "should consider" — use "must" or "must not")
- Compatible with Arc plan-schema for downstream conversion
Traditional PRD Core (25 pts)
- Clear problem/solution articulation
- Stakeholder alignment (personas, non-goals)
- Market/user context with evidence
- Measurable success criteria
Implementation Clarity (30 pts)
- Phased implementation plan with dependency order
- Exit criteria per phase
- Risk identification with mitigations
- Integration with Arc workflow (plan → spec → beads)
Completeness (20 pts)
- All required sections present and populated
- Personas with behavioral detail (not demographics-only)
- Non-goals explicitly stated
- Metrics with baselines and targets
Scoring Thresholds
- 90-100: Production-ready, can proceed directly to
/arc:core:plan - 70-89: Good, minor gaps to address before planning
- 50-69: Needs revision, significant sections incomplete
- Below 50: Major rework needed, return to clarification phase
Output Location
PRDs are written to openspec/plans/ with naming: YY-MM-DD-prd-<feature-slug>-<hash5>.md
Integration with Arc Workflow
/arc:specialized:prd <feature>
↓ writes
openspec/plans/YY-MM-DD-prd-<slug>-<hash5>.md
↓ consumed by
/arc:core:plan (creates implementation plan from PRD)
↓ continues
/arc:core:spec → /arc:core:beads → /arc:core:execute
Validation
When reviewing a PRD, check:
- All required sections present
- FR numbers are sequential and unique
- Phases are dependency-ordered
- Acceptance criteria are testable (not vague)
- Non-goals don't contradict goals
- Personas are evidence-based (not assumed)
- Score >= 70 before proceeding to
/arc:core:plan
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?