# agent-expert

A subagent that helps you design other Claude agents: structure, expertise boundaries, prompts and examples.

- subagent · `type=agent` · Claude Code · 1426 installs
- Author: davila7
- License: not stated · policy=link-only · description only; the file stays with the author
- Source: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/expert-advisors/agent-expert.md
- Updated: 2026-10-24
- HTML: https://affpapa.org/en/nevibecode/agents/agent-expert

## What it does

agent-expert is a Claude subagent for building specialized agents. It works through the domain, sets expertise boundaries, drafts a description with examples, and lays it all out as a Markdown file in the claude-code-templates format.

Inside you get templates for several agent types: technical specializations (frameworks, databases, infrastructure), domain experts (security, performance, testing), industry agents, and workflow ones like code review or architecture design. Prompt engineering gets its own treatment: how to describe competency areas, how to write examples with context, and how to state limitations so an agent stays out of someone else's lane.

It earns its place when you maintain a set of agents and want them to stop overlapping. Creation starts with domain analysis. Without a clear idea of what the agent handles and what it doesn't, there's nothing to fill the template with. The source text is truncated, so check the agent file itself for the full list of sections and format requirements.

## Install

### Claude Code (native platform)

1. **claude-code-templates**
```bash
npx claude-code-templates@latest --agent expert-advisors/agent-expert
```

2. **Manually: put the files in place**
- File: `.claude/agents/agent-expert.md` — For all projects use ~/.claude/agents/. Claude delegates by the description field or an @-mention

### Codex

**Manually: put the files in place**
- File: `.codex/agents/agent-expert.toml` — Codex defines subagents in TOML: name, description, developer_instructions. Move the Markdown text into developer_instructions by hand. For all projects use ~/.codex/agents/

### Cursor

**Manually: put the files in place**
- File: `.cursor/agents/agent-expert.md` — Cursor also reads .claude/agents/

Where a file is given instead of a command, copy the files from the source to that path: https://github.com/davila7/claude-code-templates/tree/HEAD/cli-tool/components/agents/expert-advisors/agent-expert.md

## When to use it
- write a new Claude agent in Markdown in the right format
- split expertise areas between several agents
- find examples with context for an agent description
- state limitations so an agent stays inside its domain
- pick the right agent type: technical, domain or workflow

## Example prompts
- Help me create an agent for React performance optimization
- Design an agent that covers both frontend and backend security
- Review my agent description and point out where the expertise boundaries blur

## Similar
- [h3-prompt-writing](https://affpapa.org/en/nevibecode/skills/h3-prompt-writing.md) — skill · 11600 installs
- [asd-ste100](https://affpapa.org/en/nevibecode/skills/asd-ste100.md) — skill · 7400 installs
- [skill-creator](https://affpapa.org/en/nevibecode/skills/openai-skill-creator.md) — skill · 4900 installs
- [prompt-master](https://affpapa.org/en/nevibecode/skills/prompt-master.md) — skill · 4000 installs
- [ralph-plan](https://affpapa.org/en/nevibecode/skills/ralph-plan.md) — skill · 2600 installs
- [senior-prompt-engineer](https://affpapa.org/en/nevibecode/skills/senior-prompt-engineer.md) — skill · 970 installs

From the same set: [davila7/claude-code-templates](https://affpapa.org/en/nevibecode/by/davila7/claude-code-templates.md) — 224 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