li-plan

Generates requirement-traced specifications, dependency-ordered build cards, and executor prompts from designs.

Updated May 27, 2026
One-click install
npx skills add https://github.com/jokerman89/lintel --skill li-plan-jokerman89
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: li-plan
Source: https://github.com/jokerman89/lintel/tree/main/.github/skills/li-plan
Command: npx skills add https://github.com/jokerman89/lintel --skill li-plan-jokerman89

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Teams often jump from a design or issue straight into code, losing traceability between requirements and implementation. This Skill turns a design into a structured specification, ordered build cards with acceptance criteria, and a prompt a fresh executor session can pick up cold. ## Core Features & Use Cases - Requirement-Traced Specs: Converts a design into a specification where every requirement is traceable through the plan. - Dependency-Ordered Build Cards: Produces small, sequenced build cards so work can be executed and reviewed incrementally. - Cold-Executor Prompt: Generates a self-contained prompt so a new agent session can execute the plan without prior context. - Use Case: After agreeing on a design for a new health endpoint, run the plan workflow to produce the spec and build cards, then hand the executor prompt to a fresh Copilot session to implement them. ## Quick Start Ask your Copilot agent to run /li-plan with a short description of the change, such as planning a health endpoint with a focused acceptance test.

Frequently Asked Questions about li-plan

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

FAQPage Schema
How do I turn a design into an implementation plan with GitHub Copilot?▼

Run the li-plan skill in your Copilot client with a short description of the change. It reads the Copilot adapter contract, executes the canonical plan workflow, and writes a requirement-traced spec, build cards, and an executor prompt to your repository.

What outputs does the li-plan skill generate?▼

It produces three artifacts: a specification traced back to requirements, dependency-ordered build cards with acceptance criteria, and a cold-executor prompt that a fresh agent session can use to implement the plan.

Does li-plan work outside of GitHub Copilot clients?▼

This directory is the Copilot adapter and reads the Copilot adapter contract first. The underlying canonical plan workflow lives in the shared skills directory, which other agent clients can reference through their own adapters.

Why does li-plan reference a canonical workflow file instead of containing instructions?▼

It acts as a thin adapter so the workflow logic stays in one shared source. This keeps behavior consistent across Copilot, Claude Code, and Codex users while each client applies its own tool mapping and authorization rules.

What are the limitations of the li-plan skill?▼

It depends on the canonical plan workflow and adapter contract files being present at the referenced relative paths. It contains no scripts of its own, so all behavior comes from those shared resources and the agent client's execution.