loop-executor

Execute verified plan tasks sequentially with code updates and commit logging.

2|Updated Feb 25, 2026
One-click install
npx skills add https://github.com/AhmedElhadarey/Gilfoyle --skill loop-executor
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: loop-executor
Source: https://github.com/AhmedElhadarey/Gilfoyle/tree/main/skills/loop-executor
Command: npx skills add https://github.com/AhmedElhadarey/Gilfoyle --skill loop-executor

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Automates the Execute phase of the Evaluate-Loop by implementing tasks from a verified plan, updating the plan and recording progress to ensure traceable, checkpointed execution.

Core Features & Use Cases

  • Sequential task execution: Executes each pending task in plan.md in order, respecting dependencies and task status.
  • Code generation and file updates: Writes code, creates necessary files, and updates plan.md after every completed task with a summary and commit reference.
  • Commit-logging and traceability: Logs commits at checkpoints, enabling reproducibility and auditability of the execution track.
  • Guardrails: Ensures the evaluation phase completed before execution begins and validates acceptance criteria for each task.

Quick Start

Trigger this agent with a verified plan to automatically execute tasks, write code, update plan.md after each step, and commit changes at checkpoints.

Frequently Asked Questions about loop-executor

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

FAQPage Schema
How do I automate code generation from a verified plan sequentially?

Automating sequential code generation from a verified plan involves executing tasks in order, writing code, and updating plan.md after each step. This enforces task dependencies and logs commits for reproducible execution trails.

Why does sequential task execution require updating plan.md after every step?

Updating plan.md after every step maintains a traceable execution trail during sequential task execution. This checkpointed approach records task summaries and commit metadata, ensuring reproducibility and auditability.

Can I execute plan tasks before the evaluation phase is completed?

No, you cannot execute plan tasks before the evaluation phase is completed. Guardrails ensure the evaluation phase passed before execution begins, validating acceptance criteria for each task to guarantee readiness.

How do I commit code changes at checkpoints during workflow automation?

Committing code changes at checkpoints during workflow automation happens automatically after completing tasks from a verified plan. The execution trail logs commit metadata at these checkpoints, enabling reproducible outcomes.

What are the limitations of automating plan execution without guardrails?

Automating plan execution without guardrails risks skipping the evaluation phase, executing tasks out of order, and ignoring dependencies. This breaks traceable execution trails and prevents validating acceptance criteria for each task.