mit einem Klick
ai-sdd-workflow
ai-sdd-workflow enthält 19 gesammelte Skills von ToshikiImagawa, mit Repository-Berufsabdeckung und Skill-Detailseiten auf SkillsMP.
Skills in diesem Repository
Analyze and extract requirements from use case diagram. Use when extracting UR/FR/NFR from use cases or when called by generate-prd.
Check consistency between implementation code and design documents (design), detecting discrepancies
Generate quality assurance checklists from specifications and plans with structured IDs and categories
Analyze specifications and generate clarification questions to eliminate ambiguity before implementation
Define and manage non-negotiable project principles (Constitution) and verify synchronization with other documents
Automatically executed during document updates or before implementation to check consistency between PRD ↔ *_spec.md ↔ *_design.md. Detects missing requirement ID (UR/FR/NFR) references, data model mismatches, API definition discrepancies, terminology inconsistencies, and ensures traceability between documents.
Finalize and integrate PRD from all artifacts. Use when combining use case diagrams, requirements analysis, and requirements diagrams into a complete PRD, or when called by generate-prd.
Generates complete PRD document from business requirements. Creates use case diagrams, requirements analysis (UR/FR/NFR), SysML diagrams, and complete PRD file. Use when user mentions PRD, product requirements, feature definition, requirement specification, or starting AI-SDD workflow.
Generate SysML requirements diagram from requirements analysis. Use when visualizing requirement relationships, creating traceability diagrams, or when called by generate-prd.
Generate Abstract Specification and Technical Design Document from input content
Generate use case diagram in Mermaid format from business requirements. Use when user needs to visualize actors, use cases, and system boundaries for a feature or when PRD diagrams are requested.
Execute TDD-based implementation and progressively complete checklist in tasks.md
Plan refactoring for existing features. Analyzes current implementation and creates/updates design documents with refactoring plan.
Scan existing AI-SDD documents and recommend YAML front matter additions
Automatically verify checklist items generated by /checklist command. Runs tests, linters, security scanners, and spec consistency checks to validate implementation against quality criteria.
Initialize AI-SDD workflow in the current project. Sets up CLAUDE.md and generates document templates.
Break down tasks from technical design document, generating a list of independently testable small tasks
Clean up task/ directory after implementation completion, integrating important design decisions into *_design.md before deletion
Automatically executed before implementation to analyze user instructions and detect Vibe Coding (problems where AI must infer undefined requirements due to vague instructions). Warns when detecting ambiguous expressions like 'make it nice', 'somehow', 'same as before', specification gaps, or unclear requirements, prompting for clarification.