22
GitHub Stars
3
Bundled Files
2 months ago
Catalog Refreshed
4 months ago
First Indexed
Readme & install
Copy the install command, review bundled files from the catalogue, and read any extended description pulled from the listing source.
Installation
Preview and clipboard use veilstrat where the catalogue uses aiagentskills.
npx veilstrat add skill nahisaho/musubi --skill software-developer- SKILL.md43.1 KB
- solid-principles.md8.5 KB
- test-first-workflow.md9.8 KB
Overview
This skill is a software developer agent that implements clean, maintainable, and testable code across multiple languages and frameworks. It follows SOLID principles, design patterns, and language-specific best practices to deliver high-quality implementations. It coordinates staged implementation, testing, and progress reporting for feature delivery.
How this skill works
The skill inspects requirements, design documents, and project steering files (when present) to align implementation with existing architecture and conventions. It generates code one file at a time, creates accompanying tests, enforces lint/format rules, and reports progress after each file. For larger files it splits output into parts; on completion it prepares a summary and suggests next steps like code review or test enhancements.
When to use it
- When you need a developer to implement features from requirements or design docs
- When you want code that follows SOLID, clean architecture, and language best practices
- When you require stepwise file generation with progress updates
- When tests, linting, and traceability IDs must be produced alongside code
- When you need implementation aligned with existing project steering files
Best practices
- Provide the primary English steering/docs; the agent will read them before starting if present
- Answer the agent's single-question prompts one at a time to preserve workflow and context
- Specify the target layer (frontend/backend/full-stack), tech stack, and coding conventions up front
- Co-locate requirements and API specs so the agent can extract clear acceptance criteria
- Request code review after implementation to catch architecture or integration issues
Example use cases
- Implement a user-auth feature (frontend hook, service layer, API routes, and tests) following a provided API spec
- Refactor an existing module to follow SOLID and add unit tests and linters
- Create a new backend endpoint with request validation, error handling, and integration tests
- Bootstrap feature files (types, services, hooks, components) in a feature-based project layout with progress reports
- Split large file implementations into multiple parts and produce a final summary with test coverage metrics
FAQ
Provide the feature name, implementation layer, tech stack, relevant design/requirements docs, and any project-specific coding rules.
How does the agent handle large files?
Large files are split into parts (recommended ≤300 lines per file). Each part is generated, saved, and reported sequentially to avoid context overflow.
Will tests be created with the code?
Yes. Unit and relevant integration tests are generated alongside implementation, with a target coverage and co-located test files.