From leapfrog-skill
LEAP: LLM Engineered Application Pattern. Use when the user says 'build the app', 'Read master.md', 'rewrite this project', or when you find a master.md in the project root. Generates all code from prompts, runs tests, serves the app. Can also convert existing codebases into LEAP format.
How this skill is triggered — by the user, by Claude, or both
Slash command
/leapfrog-skill:leapThis skill is limited to the following tools:
The summary Claude sees in its skill listing — used to decide when to auto-load this skill
You are operating in a LEAP project. Code is not the product — prompts, schemas, and tests are. You are the compiler.
You are operating in a LEAP project. Code is not the product — prompts, schemas, and tests are. You are the compiler.
src/ and is gitignored.The guardrails hierarchy (most to least valuable):
When you encounter a LEAP project (has master.md at root):
master.md — it contains the app description and generation instructionsschemas/ — these are the contractsparts/*/master.md has a prompt, each parts/*/schema.ts has a contractsrc/ — satisfy every contractnpm install (or equivalent)master.md has a "Running the app" section, execute itsrc/. Prompts, schemas, and tests are human-authored. You only generate code.src/ must be self-contained. No imports from outside src/. Copy type definitions inline if needed.The whole point of LEAP is that AI generates code from prompts, schemas, and tests. If you copy code from somewhere else, the experiment is invalid and the project's value is destroyed.
Forbidden:
_original/, _reference/, or any other reference folder in the projectnpm pack, git clone of the upstream, etc.)WebFetch or WebSearch to look up the implementationAllowed inputs for generation:
master.md (root and per-part)schemas/ (the contracts)tests/ (the guardrails)Why this matters: If you copy the reference, you've proven nothing — you've just laundered existing code through a prompt-shaped wrapper. LEAP's thesis is that prompts + schemas + tests are sufficient to produce correct code. Copying invalidates the experiment.
If a prompt is ambiguous: Make your best guess from the tests. If tests don't disambiguate, ask the user. Never resolve ambiguity by reading the original source.
If _original/ exists in the project: It is the user's reference, NOT yours. Treat it as if it doesn't exist during generation. You may only read it during rewrite mode setup (extracting schemas, porting tests) — never during code generation.
A LEAP project follows this structure:
master.md # Root prompt — read this first
schemas/ # Shared contracts between parts (human-authored)
parts/ # One folder per component
{name}/
master.md # Prompt for this part
schema.ts # Contract for this part
parts/ # Optional: recursive sub-parts
tests/ # Human-authored behavioral tests
src/ # GITIGNORED — you generate this
src/ is empty — generate everythingWhen the user asks to "rewrite" or "convert" an existing codebase into LEAP format:
schemas/ with type definitions covering the full public APItests/parts/ with one folder per natural componentmaster.md prompt describing what that component doesschema.ts with its specific contractmaster.md describing the project, its purpose, and how parts connectsrc/ to .gitignoremaster.md and buildThis skill is portable. The LEAP pattern works with any AI coding agent. To adopt it outside Claude Code, inject these instructions into your agent's system prompt or configuration file.
The core contract is simple:
master.mdProvides a checklist for code reviews covering functionality, security, performance, maintainability, tests, and quality. Use for pull requests, audits, team standards, and developer training.
npx claudepluginhub safitudo/leap-skill --plugin leapfrog-skill