molyanov-ai-dev
Structured feature planning, coding, testing, and review workflows
All Skills in This Repository (20)
Pure Emerald Level Indicatorsproject-knowledge
Retrieve project architecture, tech stack, and coding patterns from markdown files.
skill-test-designer
Design test scenarios for AI skills with acceptance criteria and model selections.
tech-spec-planning
Generate a technical specification document from user specs and code research.
code-writing
Guide TDD-driven code writing through preparation, implementation, and post-work phases.
post-deploy-qa
Execute Agent Verification Plans and check acceptance criteria on live environments.
project-planning
Conduct adaptive interviews to generate project documentation and technical decisions.
feature-execution
Orchestrates software feature execution by spawning AI agents in parallel waves with reviews and commits.
user-spec-planning
Generates a user-spec.md document via adaptive interview, codebase scanning, and dual validation.
pre-deploy-qa
Execute test suites and verify acceptance criteria from user and technical specifications before deployment.
methodology
Explain the AI-First development methodology and its spec-driven pipeline.
infrastructure-setup
Automate development infrastructure setup from architecture, patterns, and deployment files.
deploy-pipeline
Configure CI/CD pipelines and deployment workflows across multiple platforms.
Frequently Asked Questions
FAQPage SchemaHow to install molyanov-ai-dev?โผ
Run `npx skills add pavel-molyanov/molyanov-ai-dev --all -g -y` in your terminal to install all skills globally.
What does this framework do?โผ
It gives your coding agent a complete workflow: plan features through structured interviews, implement code and layouts, run tests, and verify results with independent reviewers.
Does it work with Claude Code and Codex?โผ
Yes. It maintains one Claude-side source of truth and automatically generates matching Codex runtime files with the included sync scripts.
How does it prevent scope creep?โผ
Every change must trace back to an approved user spec or explicit request, and reviewers only report evidence-backed findings without expanding the authorized work.
Can non-developers use this framework?โผ
Yes. You describe what you want in plain language, and the planning skill interviews you to produce a clear approved spec before any code is written.
Related Repositories in Software Engineering
View All in Software Engineeringโopenclaw
Run a personal AI assistant across your devices and chat apps
superpowers
Gives coding agents a disciplined workflow from idea to merged code
react
AI agent skills for building, testing, and porting React core