seaworthycode

The CLI distribution surface for Seaworthy. Thin wrapper around @seaworthy/core.

Usage

# Scan current directory
npx seaworthycode .

# Scan specific directory
npx seaworthycode /path/to/project

# Output JSON
npx seaworthycode --json /path/to/project

# Write HTML report
npx seaworthycode --output report.html /path/to/project

# Override tier
npx seaworthycode --tier free /path/to/project

# Use a specific provider (e.g. Kimi, DeepSeek, Ollama)
npx seaworthycode --provider kimi /path/to/project

Flags

| Flag | Description | |------|-------------| | [path] | Target directory (default: .) | | --json | Output JSON to stdout | | --output <path> | Write HTML report to file | | --tier <tier> | Override tier (free/pro) | | --provider <provider> | LLM provider (anthropic, openai, kimi, deepseek, gemini, ollama, openrouter) | | --help | Show help |

Config

Config is stored in ~/.seaworthy/config with 0600 permissions.

{
  "licenseKey": "your-key",
  "llmProvider": "anthropic",
  "llmApiKey": "sk-...",
  "llmModel": "claude-3-5-sonnet-latest",
  "llmApiUrl": "https://api.anthropic.com"
}

Provider presets

Vibe-coder friendly aliases so you don't need to remember base URLs:

| Provider | Preset | Default model | |----------|--------|---------------| | Anthropic | anthropic | claude-3-5-sonnet-latest | | OpenAI | openai | gpt-4o-mini | | Kimi (Moonshot) | kimi / moonshot | kimi-k2-5 | | DeepSeek | deepseek | deepseek-chat | | Google Gemini | gemini | gemini-2.0-flash | | Ollama (local) | ollama | llama3.2 | | OpenRouter | openrouter | anthropic/claude-3.5-sonnet |

Set llmModel to override the default. Set llmApiUrl only if you need a custom endpoint.

Running Tests

pnpm test