| name | brownfield-ears |
| description | Convert natural language requirements into EARS (Easy Approach to Requirements Syntax) format structured documents. Use when users need to clarify ambiguous requirements, formalize feature requests, or prepare high-quality input for SDD workflow. Triggers on "EARS", "requirements format", "convert requirements", "structure requirements". |
EARS Requirements Conversion
Convert natural language requirements into EARS format structured documents.
What is EARS?
EARS (Easy Approach to Requirements Syntax) uses fixed sentence templates to:
- Eliminate ambiguity in requirements
- Ensure testability
- Improve AI comprehension accuracy
Quick Start
- Provide a natural language requirement description
- This skill converts it to EARS format
- Output is saved to
.docs/EARS/[NNN]-[feature-name].md
- Use the EARS document with
/speckit.specify for better specs
EARS Patterns
| Pattern | Template | Use Case |
|---|
| Ubiquitous | The system shall <function> | Always-required functionality |
| Event-Driven | When <trigger>, the system shall <response> | Triggered actions |
| State-Driven | If <state>, then the system shall <behavior> | State-dependent behavior |
| Optional | Where <condition>, the user can <operation> | User-optional features |
| Complex | When <trigger>, and <state>, the system shall <response> | Multi-condition triggers |
Output Location
[Project]/
├── .docs/EARS/ # EARS documents (NOT in .specify/)
│ ├── 001-user-auth.md
│ └── 002-payment.md
├── .specify/ # SDD config (separate)
└── specs/ # Feature specs (separate)
Workflow
[Natural Language] → /brownfield-ears → [EARS Doc] → /speckit.specify → [Spec]
Conversion Process
- Parse input - Extract actors, actions, conditions, data
- Categorize - Map to EARS patterns
- Number - Assign IDs:
[PREFIX]-[TYPE]-[SEQ] (e.g., AUTH-EV-001)
- Generate - Create structured document
Reference
