exec-plan

Creates or resumes structured ExecPlan documents to guide feature delivery end-to-end.

Updated Apr 29, 2025
One-click install
npx skills add https://github.com/NeonTowel/oh-my-taskfiles --skill exec-plan
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: exec-plan
Source: https://github.com/NeonTowel/oh-my-taskfiles/tree/main/dotfiles/crush/skills/exec-plan
Command: npx skills add https://github.com/NeonTowel/oh-my-taskfiles --skill exec-plan

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill provides a self-contained ExecPlan document that guides a feature's delivery from design to verification, enabling novices to implement work end-to-end within the repository.

Core Features & Use Cases

  • Self-contained planning: A living document that records goals, milestones, decisions, and test criteria in a single file.
  • Guided collaboration: Facilitates back-and-forth clarification and iteration between a product owner and developer while keeping all context in one place.
  • Observability & validation: Captures progress through milestones, a decision log, and outcomes so results are verifiable.

Quick Start

To begin, create a new ExecPlan Markdown file under .ai/plans/ (for example .ai/plans/feature-x.md) and start documenting the feature in the sections described. If you need to store temporary research or clones, put them in .ai/plans/tmp (gitignored).

Frequently Asked Questions about exec-plan

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

FAQPage Schema
How do I create a living document for feature delivery that tracks milestones and decisions?

To create a living document for feature delivery, you generate a structured ExecPlan Markdown file that records goals, milestones, a decision log, and outcomes in a single self-contained file. This document guides a coding agent to deliver a feature end-to-end while keeping all context in one place.

What is the best way to guide a coding agent to implement a feature end-to-end?

The best way to guide a coding agent end-to-end is using an ExecPlan, a self-contained document that defines explicit terms and observable outcomes. It enforces structured sections for progress, surprises, decisions, and validation steps to ensure verifiable feature delivery.

How do I resume an ongoing feature plan with a coding agent after clarification?

You resume an ongoing feature plan by reopening the existing ExecPlan document, which facilitates back-and-forth clarification and iteration between a product owner and developer. The living document retains all prior context, milestones, and decisions for seamless continuation.

Where should I store temporary research files when planning feature delivery?

You should store temporary research files or clones in the .ai/plans/tmp directory, which is gitignored. The actual ExecPlan living document itself should be created as a Markdown file under the .ai/plans/ directory.

Does feature planning with an ExecPlan require a specific framework or dependencies?

Feature planning with an ExecPlan requires no specific frameworks or dependencies. It relies on creating a self-contained Markdown document that captures verifiable outcomes, milestones, and a decision log to guide the coding agent.

What sections are required in an ExecPlan living document?

An ExecPlan living document requires specific sections for Progress, Surprises & Discoveries, Decision Log, and Outcomes. These sections enforce observability and validation by capturing milestones and clearly defined observable outcomes.