nextvm docs

Generate Markdown reference docs from every discovered module.

Synopsis

nextvm docs [--out <dir>]

Options

Option	Default	Description
--out <dir>	docs/modules	Output directory

What it does

For every module in modules/*, the command emits a Markdown file under <out>/<short-name>.md containing:

• Description (from package.json)
• Version
• Declared dependencies
• RPC procedures with names + types + Zod input shapes (regex parsed)
• Locale key lists (per locale)

The generated files are intentionally hand-readable so you can drop them into a VitePress / Astro Starlight site, or commit them as plain Markdown.

Example

nextvm docs

ℹ Generating docs for 6 module(s)
✓ @nextvm/banking → docs/modules/banking.md
✓ @nextvm/housing → docs/modules/housing.md
✓ @nextvm/inventory → docs/modules/inventory.md
✓ @nextvm/jobs → docs/modules/jobs.md
✓ @nextvm/player → docs/modules/player.md
✓ @nextvm/vehicle → docs/modules/vehicle.md
✓ Wrote 6 doc file(s) to docs/modules/

Sample output

# @nextvm/banking

NextVM banking module — accounts, transfers, audit trail

**Version:** 0.0.1

## Dependencies

- `player`

## RPC Procedures

| Name | Type | Input |
| --- | --- | --- |
| `getMyBalance` | `query` | _(none)_ |
| `transfer` | `mutation` | `z.object({ toCharId: z.number()... })` |

## Locales

### `en` (4 keys)
- `banking.transfer_success`
- `banking.transfer_failed`
- ...

Limitations

• The RPC parser is regex-based — it handles the common procedure.input(z.object({ ... })) pattern but doesn't understand schemas built dynamically at runtime.
• For full TSDoc-rendered API references, drop the generated Markdown into a VitePress site (which is exactly what NextVM's own docs site does).

See also

• [com/nextvm-official/nextvm/tree/main/docs/concept)
• Modules section