# subagent-task-execution

subagent-task-execution runs a plan task by task, handing each one to its own subagent and checking the result twice before moving on.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 172 installs · ★ 4 repo stars
- Author: davidkiss
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/davidkiss/smart-ai-skills/tree/HEAD/skills/subagent-task-execution/SKILL.md
- Updated: 2026-05-08
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/subagent-task-execution

## What it does

Point the skill at a plan or a task breakdown and it works through the list in order. It reads the document end to end, pulls out every task with its context and keeps them in a todo list via `TodoWrite`. Each task then goes to a separate implementer subagent.

Before dispatch, the skill works out which other skills the task needs and passes them along with the description. Any questions from the subagent get answered before work starts. Two reviews follow: one subagent checks the result against the task requirements, another judges quality. Open findings block the next task.

A last pass looks over the whole body of work for consistency. Progress tracking depends on `TodoWrite`; without that tool in the session there is nowhere to keep the list, so the order ends up in your head.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add davidkiss/smart-ai-skills --skill subagent-task-execution -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add davidkiss/smart-ai-skills --skill subagent-task-execution -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add davidkiss/smart-ai-skills --skill subagent-task-execution -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install davidkiss/smart-ai-skills skills/subagent-task-execution
```

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

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add davidkiss/smart-ai-skills --skill subagent-task-execution -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/davidkiss/smart-ai-skills --path skills/subagent-task-execution
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add davidkiss/smart-ai-skills --skill subagent-task-execution -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/davidkiss/smart-ai-skills/tree/HEAD/skills/subagent-task-execution/SKILL.md

## When to use it
- work through a long plan one task at a time
- stop context from leaking between tasks
- catch a mismatch with the spec before the task closes
- keep the requirements check apart from the quality review
- bring in the right skill for each task

## Example prompts
- Take the tasks in `plan.md` and run them in order, one subagent per task.
- Build the todo list from `tasks.md` with `TodoWrite` and keep it current as you go.
- Send the result to a spec compliance review and a quality review before marking the task done.

## Similar
- [task_plan](https://affpapa.org/en/nevibecode/agents/task-plan.md) — subagent · ★ 395 repo stars
- [distributed-task-execution](https://affpapa.org/en/nevibecode/skills/distributed-task-execution.md) — skill · 134 installs
- [task-execution-engine](https://affpapa.org/en/nevibecode/skills/task-execution-engine.md) — skill · 65 installs
- [code-task-generator](https://affpapa.org/en/nevibecode/skills/code-task-generator.md) — skill · ★ 3206 repo stars
- [deferred-task-execution](https://affpapa.org/en/nevibecode/skills/deferred-task-execution.md) — skill · ★ 173 repo stars
- [task-planning](https://affpapa.org/en/nevibecode/skills/task-planning.md) — skill · 415 installs

From the same set: [davidkiss/smart-ai-skills](https://affpapa.org/en/nevibecode/by/davidkiss/smart-ai-skills.md) — 6 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
