namecast

内容来源：README.md（说明文档） · 原始地址 · 查看安装指南

原始内容

# Namecast

AI-powered brand name oracle. Generate, filter, and evaluate brand names with domain checks, social handles, similar company analysis, and dynamic persona perception forecasting.

## Features

- **Name Generation**: AI generates brand name candidates based on your project description
- **Domain Availability**: Checks .com, .io, .co, .ai, .app domains via WHOIS
- **Social Handle Checks**: Twitter/X and GitHub availability
- **Similar Company Search**: Finds existing companies with similar names
- **Pronunciation Analysis**: Syllable count, spelling difficulty, phonetic clarity
- **International Safety**: Checks for problematic meanings in 7 major languages
- **Dynamic Persona Perception**: AI role-plays as your target audience to forecast brand perception

## Installation

```bash
pip install namecast
```

Or from source:

```bash
git clone https://github.com/MaxGhenis/namecast
cd namecast
pip install -e ".[dev]"
```

Requires an Anthropic API key:

```bash
export ANTHROPIC_API_KEY=your-key-here
```

## Usage

### CLI

**Find names for a new project:**

```bash
namecast find "A SaaS tool for tracking carbon emissions"
```

**Include your own ideas:**

```bash
namecast find "Dog walking app" --ideas Waggle --ideas PupPath
```

**Evaluate a specific name:**

```bash
namecast eval Acme
```

**Evaluate with mission context:**

```bash
namecast eval Cloudify --mission "Cloud infrastructure automation for startups"
```

**Compare multiple names:**

```bash
namecast eval --compare Acme Globex Initech
```

### Claude Code Plugin

If you use [Claude Code](https://claude.ai/claude-code), Namecast is available as a plugin:

**Find names:**

```
/namecast:find A marketplace connecting local farmers with restaurants --ideas FarmLink,HarvestHub
```

**Evaluate a name:**

```
/namecast:evaluate Cloudify --mission Cloud infrastructure automation
```

The Claude Code plugin runs natively (no API calls) and is free with Claude Max.

## Methodology

### Scoring System

Names are scored on a 100-point scale:

| Category | Points | Description |
|----------|--------|-------------|
| Domain Availability | 40 | .com = 40pts, .io only = 25pts |
| Persona Perception | 40 | Average of memorability + professionalism ratings |
| Linguistic Quality | 20 | Pronunciation ease, international safety, uniqueness |

### Dynamic Persona Generation

Unlike static persona sets, Namecast generates **personas tailored to your specific project**. When you provide a company description, the AI identifies 5 relevant personas who would interact with your brand:

- **Primary customers/users** - Who will use the product?
- **Decision makers** - Who approves purchases?
- **Investors** - Who might fund the company?
- **Industry experts** - Who analyzes this space?
- **End consumers** - Who benefits ultimately?

For example, a B2B SaaS might get personas like:
- CFO (50s, enterprise, values ROI)
- Startup founder (30s, fast-mover, values innovation)
- Developer (25, technical, values ease of use)
- VC partner (40s, evaluates hundreds of startups)
- Industry analyst (45, writes about the space)

While a consumer app might get:
- College student (20, social media native)
- Working parent (35, values convenience)
- Small business owner (40, budget-conscious)

Each persona then evaluates your name on:
- **Memorability** (1-10): Will they remember it?
- **Professionalism** (1-10): Does it seem credible?
- **Trust**: Would they use a company with this name?
- **Industry guess**: What sector do they assume?
- **Gut reaction**: One-sentence impression

### Domain Checking

Checks WHOIS data for:
- `.com` (weighted highest)
- `.io` (popular for tech)
- `.co`
- `.ai`
- `.app`

Names where both .com and .io are taken are filtered out in the `find` workflow.

### Social Handle Verification

Checks availability on:
- Twitter/X
- GitHub

### Similar Company Analysis

Web searches for:
- Exact name matches
- Companies in similar industries
- Potential trademark conflicts

**Note:** This is informational only. Always consult a trademark attorney before finalizing your brand name.

### International Safety

Checks if the name has problematic meanings in:
- Spanish
- French
- German
- Mandarin
- Japanese
- Portuguese
- Arabic

### Pronunciation Analysis

Evaluates:
- **Syllable count**: Shorter is generally better
- **Spelling difficulty**: Easy/Medium/Hard
- **Phonetic clarity**: Would someone spell it correctly after hearing it?

## Architecture

```
namecast/
├── namecast/
│   ├── cli.py          # Click CLI commands
│   ├── evaluator.py    # BrandEvaluator and NamecastWorkflow
│   ├── perception.py   # Dynamic persona generation and analysis
│   ├── api.py          # FastAPI server (optional)
│   └── tests/          # pytest test suite
├── .claude-plugin/
│   ├── plugin.json     # Claude Code plugin manifest
│   ├── marketplace.json
│   └── commands/
│       ├── find.md     # /find command (native execution)
│       └── evaluate.md # /evaluate command (native execution)
└── src/                # React frontend (optional)
```

### CLI vs Plugin Alignment

Both the CLI and Claude Code plugin use the same methodology:

| Feature | CLI | Claude Code Plugin |
|---------|-----|-------------------|
| Dynamic personas | `perception.py` | Native Claude roleplay |
| Domain checks | `python-whois` | WebFetch to who.is |
| Social checks | `httpx` | WebFetch |
| Similar companies | `httpx` | WebSearch |
| API cost | Anthropic API credits | Free (Claude Max) |

The key difference: CLI calls the Anthropic API (costs credits), while the Claude Code plugin uses Claude's native capabilities (free with Claude Max subscription).

## Development

```bash
# Install dev dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run linting
ruff check namecast/
```

## License

MIT

## Disclaimer

Namecast provides general information only and does not constitute legal advice. Domain availability and trademark searches are informational. Always consult a trademark attorney before finalizing your brand name.

Namecast

AI-powered brand name oracle. Generate, filter, and evaluate brand names with domain checks, social handles, similar company analysis, and dynamic persona perception forecasting.

Features

Name Generation: AI generates brand name candidates based on your project description
Domain Availability: Checks .com, .io, .co, .ai, .app domains via WHOIS
Social Handle Checks: Twitter/X and GitHub availability
Similar Company Search: Finds existing companies with similar names
Pronunciation Analysis: Syllable count, spelling difficulty, phonetic clarity
International Safety: Checks for problematic meanings in 7 major languages
Dynamic Persona Perception: AI role-plays as your target audience to forecast brand perception

Installation

pip install namecast

Or from source:

git clone https://github.com/MaxGhenis/namecast
cd namecast
pip install -e ".[dev]"

Requires an Anthropic API key:

export ANTHROPIC_API_KEY=your-key-here

Usage

CLI

Find names for a new project:

namecast find "A SaaS tool for tracking carbon emissions"

Include your own ideas:

namecast find "Dog walking app" --ideas Waggle --ideas PupPath

Evaluate a specific name:

namecast eval Acme

Evaluate with mission context:

namecast eval Cloudify --mission "Cloud infrastructure automation for startups"

Compare multiple names:

namecast eval --compare Acme Globex Initech

Claude Code Plugin

If you use Claude Code, Namecast is available as a plugin:

Find names:

/namecast:find A marketplace connecting local farmers with restaurants --ideas FarmLink,HarvestHub

Evaluate a name:

/namecast:evaluate Cloudify --mission Cloud infrastructure automation

The Claude Code plugin runs natively (no API calls) and is free with Claude Max.

Methodology

Scoring System

Names are scored on a 100-point scale:

Category	Points	Description
Domain Availability	40	.com = 40pts, .io only = 25pts
Persona Perception	40	Average of memorability + professionalism ratings
Linguistic Quality	20	Pronunciation ease, international safety, uniqueness

Dynamic Persona Generation

Unlike static persona sets, Namecast generates personas tailored to your specific project. When you provide a company description, the AI identifies 5 relevant personas who would interact with your brand:

Primary customers/users - Who will use the product?
Decision makers - Who approves purchases?
Investors - Who might fund the company?
Industry experts - Who analyzes this space?
End consumers - Who benefits ultimately?

For example, a B2B SaaS might get personas like:

CFO (50s, enterprise, values ROI)
Startup founder (30s, fast-mover, values innovation)
Developer (25, technical, values ease of use)
VC partner (40s, evaluates hundreds of startups)
Industry analyst (45, writes about the space)

While a consumer app might get:

College student (20, social media native)
Working parent (35, values convenience)
Small business owner (40, budget-conscious)

Each persona then evaluates your name on:

Memorability (1-10): Will they remember it?
Professionalism (1-10): Does it seem credible?
Trust: Would they use a company with this name?
Industry guess: What sector do they assume?
Gut reaction: One-sentence impression

Domain Checking

Checks WHOIS data for:

.com (weighted highest)
.io (popular for tech)
.co
.ai
.app

Names where both .com and .io are taken are filtered out in the find workflow.

Social Handle Verification

Checks availability on:

Twitter/X
GitHub

Similar Company Analysis

Web searches for:

Exact name matches
Companies in similar industries
Potential trademark conflicts

Note: This is informational only. Always consult a trademark attorney before finalizing your brand name.

International Safety

Checks if the name has problematic meanings in:

Spanish
French
German
Mandarin
Japanese
Portuguese
Arabic

Pronunciation Analysis

Evaluates:

Syllable count: Shorter is generally better
Spelling difficulty: Easy/Medium/Hard
Phonetic clarity: Would someone spell it correctly after hearing it?

Architecture

namecast/
├── namecast/
│   ├── cli.py          # Click CLI commands
│   ├── evaluator.py    # BrandEvaluator and NamecastWorkflow
│   ├── perception.py   # Dynamic persona generation and analysis
│   ├── api.py          # FastAPI server (optional)
│   └── tests/          # pytest test suite
├── .claude-plugin/
│   ├── plugin.json     # Claude Code plugin manifest
│   ├── marketplace.json
│   └── commands/
│       ├── find.md     # /find command (native execution)
│       └── evaluate.md # /evaluate command (native execution)
└── src/                # React frontend (optional)

CLI vs Plugin Alignment

Both the CLI and Claude Code plugin use the same methodology:

Feature	CLI	Claude Code Plugin
Dynamic personas	`perception.py`	Native Claude roleplay
Domain checks	`python-whois`	WebFetch to who.is
Social checks	`httpx`	WebFetch
Similar companies	`httpx`	WebSearch
API cost	Anthropic API credits	Free (Claude Max)

The key difference: CLI calls the Anthropic API (costs credits), while the Claude Code plugin uses Claude's native capabilities (free with Claude Max subscription).

Development

# Install dev dependencies
pip install -e ".[dev]"

# Run tests
pytest

# Run linting
ruff check namecast/

License

MIT

Disclaimer

Namecast provides general information only and does not constitute legal advice. Domain availability and trademark searches are informational. Always consult a trademark attorney before finalizing your brand name.