Agent skill
speckit-specify
Create or update feature specification from natural language feature description
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/other/speckit-specify-kyteapp-growth-agents-and-sk
SKILL.md
Speckit-Specify: Feature Specification Creation
Purpose
Create a formal, technology-agnostic specification that describes WHAT users need and WHY it matters, without prescribing HOW to implement it.
Prerequisites
- Completed brainstorming session (use
brainstormingskill first) - Clear understanding of user requirements and constraints
What This Skill Does
- Generates feature branch and directory structure
- Creates comprehensive
spec.mdfollowing template - Validates specification quality
- Resolves remaining clarifications (max 3)
- Ensures specification is ready for technical planning
Key Principles
Focus on WHAT and WHY
- Describe user capabilities and business value
- Avoid implementation details (no languages, frameworks, APIs)
- Written for business stakeholders, not developers
Technology-Agnostic Success Criteria
- Measurable outcomes from user/business perspective
- No mention of frameworks, databases, or tools
- Examples: "Users can complete checkout in <3 minutes", "System supports 10,000 concurrent users"
Quality Over Speed
- Make informed guesses for reasonable defaults
- Only mark [NEEDS CLARIFICATION] for critical decisions (max 3)
- Validate against quality checklist before completion
Template Location
IMPORTANT: The specification template is located at:
.github/skills/speckit-specify/spec-template.md
This template MUST be loaded and used as the base structure for all specification creation.
Execution Flow
All execution logic is now contained within this skill. The skill handles:
- Feature branch creation and numbering
- Spec template loading and filling
- Quality validation
- Clarification handling
Quick Summary
- Parse user description from arguments
- Generate branch name (2-4 words, action-noun format)
- Check existing branches (remote, local, specs directories)
- Run create-new-feature.sh with next available number
- Load spec template from
.github/skills/speckit-specify/spec-template.md - Fill specification with concrete details from brainstorming
- Validate quality against checklist criteria
- Resolve clarifications (max 3, with suggested answers)
- Report completion with branch, spec path, and readiness
Specification Quality Validation
After writing the spec, validate against these criteria:
Content Quality
- No implementation details (languages, frameworks, APIs)
- Focused on user value and business needs
- Written for non-technical stakeholders
- All mandatory sections completed
Requirement Completeness
- No [NEEDS CLARIFICATION] markers remain
- Requirements are testable and unambiguous
- Success criteria are measurable
- Success criteria are technology-agnostic (no implementation details)
- All acceptance scenarios are defined
- Edge cases are identified
- Scope is clearly bounded
- Dependencies and assumptions identified
Feature Readiness
- All functional requirements have clear acceptance criteria
- User scenarios cover primary flows
- Feature meets measurable outcomes defined in Success Criteria
- No implementation details leak into specification
Clarification Handling
If [NEEDS CLARIFICATION] markers exist (max 3):
- Extract markers from spec
- Limit to 3 most critical (scope → security → UX → technical details)
- Present as Q&A format:
- Recommended: Option X -
- Table with options A-E
- User can say "yes"/"recommended" to accept suggestion
- Update spec with user's answers
- Re-validate quality after clarifications resolved
Success Indicators
Specification is ready when:
- ✅ Zero implementation details present
- ✅ All success criteria are measurable and technology-agnostic
- ✅ No [NEEDS CLARIFICATION] markers remain
- ✅ All quality checklist items pass
- ✅ Requirements are testable and unambiguous
- ✅ Written for business stakeholders (no technical jargon)
Output
specs/N-feature-name/
├── spec.md # Feature specification (WHAT & WHY)
└── checklists/
└── requirements.md # Quality validation checklist
Next Step
After specification is complete:
- Option A: Use
speckit-planto create technical plan - Option B: Use
speckit-checklistto generate domain-specific quality checks first
Common Mistakes
❌ Implementation Details in Spec
Wrong: "Use React hooks for state management" Right: "Users can update their profile information"
❌ Technology-Specific Success Criteria
Wrong: "API response time is under 200ms" Right: "Users see search results instantly"
❌ Too Many Clarifications
Wrong: 10 [NEEDS CLARIFICATION] markers Right: Max 3, make informed guesses for the rest
❌ Vague Requirements
Wrong: "System should be fast and secure" Right: "Users can log in within 2 seconds, all data encrypted at rest and in transit"
Related Skills
- brainstorming - REQUIRED before this skill (explore requirements)
- speckit-plan - Next step after specification (design technical approach)
- speckit-checklist - Generate quality validation checklists
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?