| name | read-ticket |
| description | Stage 1: Ticket Intake - Understand what to build before coding begins.
Identifies architecture layers, anticipates classes, determines test types, flags ambiguities.
Trigger: "read ticket CA-123", "analyze ticket", "what does CA-456 require"
|
| disable-model-invocation | false |
Read Ticket Skill
Verb: Understand what to build.
Agent's first action when assigned a story. Output primes plan-implementation and subsequent coding skills.
Input
ticket_id: YouTrack ID (e.g., "CA-123")
output_format (optional): structured or conversational
Workflow
-
Fetch ticket via mcp__youtrack__get_issue(issueId: ticket_id)
- Extract: summary, description, custom fields, comments
- Handle errors: ticket not found, missing description
- Fallback: If YouTrack MCP unavailable, ask user to paste ticket content directly
-
Identify layers touched (presentation, domain, data, integration)
- UI changes/screens → presentation
- Business logic/rules → domain
- API/DB/cache → data
- New SDK → integration (ask user to confirm)
-
Anticipate classes (rough scope, not final naming)
- Presentation: {Feature}Page, {Feature}Bloc/Event/State
- Domain: {Verb}{Noun}UseCase, {Noun}Repository (interface)
- Data: Remote/Local{Noun}DataSource, {Noun}RepositoryImpl
-
Determine test types
- Unit tests: always
- Widget tests: if presentation touched
- Golden tests: if layout-sensitive UI
- Mutation tests: if 4+ conditional branches (per Mutation Testing:
skills/rules/13-mutation-testing.md)
- Accessibility tests: if critical flow
-
Flag ambiguities
- Missing: user flow, edge cases, data source, design refs
- Unclear: acceptance criteria, conflicting requirements
- Surface to developer before
plan-implementation
-
Output analysis
- Layers touched + next skills needed
- Anticipated classes
- Required test types
- Ambiguities + clarifying questions
- Estimated complexity
Output Format
Conversational (default):
📋 Ticket CA-123: Add project estimation page
Layers: Presentation + Domain
Classes: EstimationPage, EstimationBloc, GetEstimationsUseCase
Tests: Unit + Widget + Golden
⚠️ Ambiguities:
- No API endpoint specified
- No error handling mentioned
**Next:** Clarify ambiguities, then plan the implementation using the plan-implementation skill
References
- Next skill:
skills/plan-implementation/SKILL.md