# delegated-development

The delegated-development skill runs a development plan through fresh subagents, with a specification check and a code-quality review after each task.

- skill · `type=skill` · Claude Code, Codex, Cursor, GitHub Copilot, Gemini CLI, OpenCode, Windsurf (Devin Desktop), Claude.ai · 142 installs · ★ 50 repo stars
- Author: qodex-ai
- License: LicenseRef-Proprietary · policy=link-only · description only; the file stays with the author
- Source: https://github.com/qodex-ai/ai-agent-skills/tree/HEAD/skills/delegated-development/SKILL.md
- Updated: 2026-01-05
- Security check: automated, 2026-11-16 (text) — no signs of malicious code found.
- HTML: https://affpapa.org/en/nevibecode/skills/delegated-development

## What it does

delegated-development turns an implementation plan into a sequence of separate tasks. A fresh subagent handles each one with its full task text and context, then implements the change, runs tests, reviews its own work, and commits the result.

Each completed task goes through two reviews. The first checks whether the implementation follows the specification; the second looks at code quality. A failed review sends the task back for fixes and another review. Once all tasks are done, a final reviewer checks the whole implementation, followed by `superpowers:finishing-a-development-branch`.

The workflow assumes a prepared plan, mostly independent tasks, and one active session. It is the wrong fit for tightly coupled work or a plan that still needs to be developed. Its prompt templates are `./implementer-prompt.md`, `./spec-reviewer-prompt.md`, and `./code-quality-reviewer-prompt.md`.

## Install

### Claude Code (native platform)

1. **Skills CLI (Vercel)**
```bash
npx skills add qodex-ai/ai-agent-skills --skill delegated-development -a claude-code
```

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

### Codex

1. **Skills CLI (Vercel)**
```bash
npx skills add qodex-ai/ai-agent-skills --skill delegated-development -a codex
```

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

### Cursor

1. **Skills CLI (Vercel)**
```bash
npx skills add qodex-ai/ai-agent-skills --skill delegated-development -a cursor
```

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

### GitHub Copilot

1. **GitHub CLI**
```bash
gh skill install qodex-ai/ai-agent-skills skills/delegated-development
```

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

### Gemini CLI

1. **Skills CLI (Vercel)**
```bash
npx skills add qodex-ai/ai-agent-skills --skill delegated-development -a gemini-cli
```

2. **Platform CLI**
```bash
gemini skills install https://github.com/qodex-ai/ai-agent-skills --path skills/delegated-development
```

### Windsurf (Devin Desktop)

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

### OpenCode

**Skills CLI (Vercel)**
```bash
npx skills add qodex-ai/ai-agent-skills --skill delegated-development -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/qodex-ai/ai-agent-skills/tree/HEAD/skills/delegated-development/SKILL.md

## When to use it
- split an implementation plan into separate tasks
- give each task to a fresh subagent
- check changes against the specification
- send failed reviews back for correction
- review the completed implementation as a whole

## Example prompts
- Execute `docs/plans/feature-plan.md` with a fresh subagent for each task and track the work in TodoWrite.
- Review every completed task for specification compliance first, then for code quality, and repeat the review after fixes.
- Take the independent tasks from this plan, implement them in sequence, and run a final review of the whole implementation.

## Similar
- [apex](https://affpapa.org/en/nevibecode/skills/melvynx-apex.md) — skill · 669 installs
- [loop-verifier](https://affpapa.org/en/nevibecode/skills/loop-verifier.md) — skill · ★ 11819 repo stars
- [constraint-driven-development](https://affpapa.org/en/nevibecode/skills/constraint-driven-development.md) — skill · 49100 installs
- [nanostack](https://affpapa.org/en/nevibecode/plugins/nanostack.md) — plugin · ★ 209 repo stars
- [gravitas](https://affpapa.org/en/nevibecode/skills/gravitas.md) — skill · 6800 installs
- [blueprint-mode-codex](https://affpapa.org/en/nevibecode/agents/blueprint-mode-codex.md) — subagent · 201 installs

From the same set: [qodex-ai/ai-agent-skills](https://affpapa.org/en/nevibecode/by/qodex-ai/ai-agent-skills.md) — 42 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
