What kind of verification does it perform?

The skill runs the full test suite, executes build commands, performs linting, and dispatches a specialized code-reviewer agent before finalizing a plan.

How does the skill track task completion?

It manages a plan file, marking tasks as IN_PROGRESS when they start and as [x] COMPLETED once they pass all verification steps like tests and linting.

How does this skill optimize agent context?

It groups tasks targeting the same directory or feature so a single agent can handle them without needing to re-explore the codebase for every individual task.

What happens if an agent encounters an error during execution?

The skill features auto-recovery; it attempts to fix failures with a dedicated agent. If the same error occurs twice, it stops and requests user intervention.

Can it run multiple tasks at the same time?

Yes, the skill identifies independent subsystems and dispatches agents to work on separate groups in parallel, significantly reducing total execution time.

Intelligent Plan Execution

Name: Intelligent Plan Execution
Author: rileyhilliard

byrileyhilliard

•

Productivity & Workflow

Executes complex implementation plans by intelligently grouping related tasks to optimize context sharing and parallelize independent subsystems.

The Intelligent Plan Execution skill streamlines the implementation phase of software development by transforming static plans into actionable, parallelized workflows. By grouping related tasks by directory or domain, it ensures that agents maintain deep context and reduce redundant codebase exploration, while simultaneously dispatching independent tasks to separate agents for maximum speed. It includes built-in progress tracking, automated verification through tests and builds, and self-recovery mechanisms that attempt to resolve failures automatically before escalating to the user.

Key Features

01Automated Verification: Integrated test, build, and linting checks ensure code quality at every step.

02Smart Task Grouping: Automatically clusters tasks by directory or domain to minimize context switching.

03Self-Recovery Logic: Employs dedicated fix agents to resolve failures without manual intervention.

0462 GitHub stars

05Real-time Progress Tracking: Updates plan files dynamically to reflect in-progress and completed tasks.

06Parallel Execution: Dispatches independent subsystems to multiple agents for faster implementation.

Use Cases

01Implementing multi-module feature updates across distinct subsystems like auth, billing, and UI.

02Automating the end-to-end development lifecycle from initial coding to final code review and verification.

03Executing large-scale refactors where changes are spread across many related files in specific directories.

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add rileyhilliard/claude-essentials executing-plans

For use in Claude.ai and ChatGPT

Download Skill