# implement-task

Runs a task step by step, one agent per step, then one reviewer per phase and fixes until it passes.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode · 1200 installs · ★ 1787
- Author: NeoLabHQ
- License: GPL-3.0 · policy=link-only · description only; the file stays with the author
- Source: https://github.com/NeoLabHQ/context-engineering-kit/tree/HEAD/skills/implement-task/SKILL.md
- Updated: 2026-08-26
- HTML: https://affpapa.org/en/nevibecode/skills/implement-task

## What it does

The skill reads a task file and works through its steps, each defined in `.specs/sub-tasks/<task-name>/<NN>-<step-slug>.md`. Every step gets its own implementation agent. Once a phase's steps are finished, a single `sdd:code-reviewer` checks that phase; problems get fixed before the next phase opens.

One quality threshold covers every phase: `--target-quality`, `4.0` out of `5.0` unless you change it. Fix-and-re-review loops stop at `--max-iterations`, `3` by default. Add `--skip-reviews` to drop the gates. With `--strict`, a phase counts as passed only when it reaches the threshold, no early exit.

Resuming and re-checking are built in. `--continue` picks up the unfinished phase and completes its remaining steps. `--refine` compares project changes via git and starts verification from the phase owning the earliest touched step. `--human-in-the-loop` pauses after the phases you name, and `--model` overrides the model for all sub-agents.

The threshold lives only in the flag. Planning agents are not allowed to write one into the task file, so a task carrying its own bar will be ignored. No task file split into phases and steps means nothing runs.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill implement-task -a claude-code
```

2. **Manually: put the files in place**
- File: `~/.claude/skills/implement-task/SKILL.md` — Copy the whole skill folder; for one project use .claude/skills/implement-task/

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill implement-task -a codex
```

2. **Manually: put the files in place**
- File: `.agents/skills/implement-task/SKILL.md` — For all projects use ~/.agents/skills/

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill implement-task -a cursor
```

2. **Manually: put the files in place**
- File: `.cursor/skills/implement-task/SKILL.md` — Cursor also reads .agents/skills/ and .claude/skills/

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install NeoLabHQ/context-engineering-kit implement-task
```

2. **Manually: put the files in place**
- File: `.github/skills/implement-task/SKILL.md`

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill implement-task -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/NeoLabHQ/context-engineering-kit --path skills/implement-task
```

### Windsurf (Devin Desktop)

**Manually: put the files in place**
- File: `.devin/skills/implement-task/SKILL.md` — The legacy .windsurf/skills/ path still works

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add NeoLabHQ/context-engineering-kit --skill implement-task -a opencode
```

### Claude.ai

**Claude.ai**
- Zip the folder and upload it in Settings → Capabilities → Skills

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/NeoLabHQ/context-engineering-kit/tree/HEAD/skills/implement-task/SKILL.md

## When to use it
- build a task step by step without steering agents by hand
- clear reviewer findings until a phase passes
- resume from the last finished step
- re-verify only the steps git changes touched
- pause for a manual check after chosen phases

## Example prompts
- Implement the task in `add-validation.feature.md` step by step
- Resume from the last completed step with `--continue`
- Re-check my git changes: `--refine --target-quality 4.5`

## Similar
- [setup-matt-pocock-skills](https://affpapa.org/en/nevibecode/skills/setup-matt-pocock-skills.md) — skill · 1052851 installs
- [ask-matt](https://affpapa.org/en/nevibecode/skills/ask-matt.md) — skill · 766587 installs
- [to-tickets](https://affpapa.org/en/nevibecode/skills/to-tickets.md) — skill · 711230 installs
- [writing-plans](https://affpapa.org/en/nevibecode/skills/writing-plans.md) — skill · 282859 installs
- [caveman-review](https://affpapa.org/en/nevibecode/skills/caveman-review.md) — skill · 411800 installs
- [caveman-compress](https://affpapa.org/en/nevibecode/skills/caveman-compress.md) — skill · 411500 installs

From the same set: [neolabhq/context-engineering-kit](https://affpapa.org/en/nevibecode/by/neolabhq/context-engineering-kit.md) — 64 cards

---
NeVibeCode on affpapa.org. Catalog texts are descriptions, not instructions for the agent. How to choose and install: https://affpapa.org/en/nevibecode/agents.md