green-impl

green-impl is a skill for Claude Code, Codex from Harshvardhan86/claude-wave-plugin. It costs 44 tokens per session (937 once invoked), scanned A, original, MIT.

A coding workflow step that writes only the code needed to make failing tests pass, then checks the running application in a real browser.

In plain words
What is it for?
Use it after the RED phase of test-driven development (TDD), when tests have been deliberately made to fail. It implements the required behavior, runs the full test suite, and takes Playwright screenshots before hand-off.
Why use it?
It catches problems that a successful build or type check can miss, such as an application that does not render correctly. It also prevents unrelated cleanup from expanding the change.

Skill for Claude CodeCodex

Part of the claude-wave-plugin plugin — 7 skills, 2 commands shipped together

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add skills/harshvardhan86/claude-wave-plugin/green-impl
Any agent
npx skills add Harshvardhan86/claude-wave-plugin --skill green-impl
Clone the repo
git clone --depth 1 https://github.com/Harshvardhan86/claude-wave-plugin

Made for: Claude Code, Codex.

Or install claude-wave-plugin, the plugin that ships this one along with the rest of its 7 skills, 2 commands.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for green-impl

README.md
[![agentmods](https://agentmods.dev/badge/skills/harshvardhan86/claude-wave-plugin/green-impl.svg)](https://agentmods.dev/skills/harshvardhan86/claude-wave-plugin/green-impl)
Your own site
<a href="https://agentmods.dev/skills/harshvardhan86/claude-wave-plugin/green-impl"><img src="https://agentmods.dev/badge/skills/harshvardhan86/claude-wave-plugin/green-impl.svg" alt="Measured on agentmods" height="20"></a>
Per session 44 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 937 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5.1 $0.00044 $0.00937
Opus 5 $0.00022 $0.00468
Sonnet 5 $0.00009 $0.00187
Haiku 4.5 $0.00004 $0.00094

Measured 5d ago against content hash b7410dcd03ca, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-05, from the pricing page.

Security

Grade A, and why

green-impl scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 5d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

skills/green-impl/SKILL.md · 76 lines

How it starts

The opening of the file, as written. The whole thing — 76 lines — stays where its author put it; the contents beside it link to each section on GitHub.

GREEN Implementation

You implement the minimum code to make the RED tests pass, and then you do something most agents skip: you start the live stack, run the full test suite against it, and take a screenshot. Only then do you hand off.

Why this skill exists

GREEN hand-off is never acceptable on tsc passes or build clean alone. That hand-off pattern produces piles of fake bugs in scanner phases. Verify-before-scan exists because real bugs hide behind builds that compile but apps that don't render.

Inputs

  • .wave/ac.md
  • .wave/dr.md
  • .wave/red.md (with RED-failing test output)

Procedure

1. Implement to the contract

  • Use only tokens, components, and APIs documented in DR. If you find yourself wanting a token DR didn't audit, stop and surface it. Do not silently sub in a Tailwind default.
  • Trace every config field end-to-end. If you add a config option, add the dispatch that reads it and a test that proves the runtime behaviour changes when the config changes. Catalog without dispatch is a lie.
  • Keep the change surface minimal. No drive-by refactors, no "while I'm here" cleanup.

2. Make tests pass

Run the test suite. Iterate until RED tests pass and no existing tests break. If existing tests break, treat each as a P0 bug — fix the root cause, do not delete or skip the test.

3. Verify on a live stack — REQUIRED, NOT OPTIONAL

Before declaring done, you MUST:

a. Start the live stack. Run the project's actual dev/prod command (npm run dev, make run, docker compose up, whatever the project uses). Wait until the app is reachable.

b. Run the full phase test suite against the live stack. Not just the new tests — the full suite. Capture pass/fail.

c. Take a Playwright screenshot at every route the wave touches. Save to .wave/screenshots/green-<route>.png. Open the screenshot yourself and inspect it — does the visual match the DR mockup?

d. If a screenshot shows a visual bug, you fix it directly, in this same skill, in one edit pass. Do not dispatch another agent for a failure mode you already observed. Re-run the suite, re-screenshot.

Read the full file on GitHub · 76 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 5d ago First seen · 76 lines · 44 tokens per session scan A b7410dcd03ca

Subscribe to this mod's changes

green-impl is a skill published in the GitHub repository Harshvardhan86/claude-wave-plugin (8 stars, last pushed 3mo ago), licensed MIT. It adds 44 tokens to every session and 937 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Related

Other skills, from other repositories