feature-execution

Orchestrates software feature execution by spawning AI agents in parallel waves with reviews and commits.

244|75|Updated Dec 22, 2025
One-click install
npx skills add https://github.com/pavel-molyanov/molyanov-ai-dev --skill feature-execution
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: feature-execution
Source: https://github.com/pavel-molyanov/molyanov-ai-dev/tree/main/skills/feature-execution
Command: npx skills add https://github.com/pavel-molyanov/molyanov-ai-dev --skill feature-execution

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

This Skill automates the complex, multi-agent process of delivering software features from technical specification to completion, managing parallel tasks, reviews, and commits.

Core Features & Use Cases

  • Wave-based Execution: Organizes tasks into parallel execution waves, respecting dependencies.
  • Automated Review Cycles: Manages up to 3 rounds of automated code reviews with specified reviewers.
  • Commit Management: Commits code after implementation and after each review round.
  • Use Case: When a new feature is defined by user and technical specifications, this Skill orchestrates the entire development process, assigning tasks to agents, managing their execution, ensuring code quality through reviews, and committing the final code.

Quick Start

Use the feature-execution skill to execute the feature defined in work/my-new-feature/tech-spec.md.

Frequently Asked Questions about feature-execution

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I orchestrate parallel AI agents for feature development?

Feature development with parallel AI agents is orchestrated by organizing tasks into execution waves, managing dependencies, spawning specialized agents, and committing code after each wave.

What is the best way to automate code reviews during agent-driven feature execution?

Automated code reviews during feature execution are managed through up to three review rounds with specified reviewers, committing code after each iteration to ensure quality before final integration.

Can I resume feature execution if a parallel agent task fails midway?

Resuming feature execution after a failure is supported through checkpointing, which saves task status updates and allows the process to restart from the last completed wave.

How do I manage task dependencies when executing software features in parallel waves?

Task dependencies during parallel wave execution are managed by the orchestrator, which sequences waves so that dependent tasks are only assigned to agents after prerequisite tasks are completed.

When should I escalate automated feature execution to human oversight?

Escalation to human oversight during automated feature execution is triggered when the orchestrator encounters unresolved issues across automated code review cycles or agent task failures.

Does feature execution require a technical specification file to start?

Feature execution requires a technical specification file, such as a tech-spec markdown document, to define the feature scope and tasks before orchestrating the parallel agent workflow.