beavers-voice-transcript-client

Node.js client library for the Beavers Voice Transcript Foundry VTT module.

Allows external tools and AI agents to read and write Foundry VTT journal entries over a socket connection.

Requirements

• The Beavers Voice Transcript module must be installed and running in Foundry VTT.
• A Gamemaster must be connected to the Foundry instance for the API to function.
• A dedicated Bot-Control user must be configured in the module settings (see Connection Info).

Installation

npm install beavers-voice-transcript-client

Usage

import { BeaversClient } from "beavers-voice-transcript-client";

const client = new BeaversClient({
  url: "http://localhost:30000",
  userId: "<userId from module Connection Info>",
  password: "<password from module Connection Info>",
});

await client.connect();

const journals = await client.listJournals();
const entry = await client.readJournal("My Journal");

await client.writeJournal({ name: "My Journal", folder: "My Folder" });
await client.writeJournalPage("My Journal", { name: "Page 1", text: { content: "<p>Hello</p>" } });

await client.disconnect();

API

new BeaversClient(options)

connect() → Promise<void>

Authenticates with Foundry and opens the socket connection. Must be called before any API method.

disconnect() → Promise<void>

Closes the socket connection.

listJournals(folder?) → Promise<object[]>

Lists journal entries and subfolders. Omit folder to list from the root.

readJournal(identifier) → Promise<object>

Reads a journal entry by name or ID.

writeJournal(data: JournalData) → Promise<object>

Creates or updates a journal entry. Matches an existing entry by id or name; creates a new one if neither matches.

interface JournalData {
  id?:        string;           // update by ID
  name?:      string;           // update/create by name
  folder?:    string;           // folder name or ID; created automatically if missing
  pages?:     JournalPageData[]; // created atomically on first creation only
  ownership?: Record<string, number>; // e.g. { default: 0 }
  flags?:     Record<string, unknown>;
}

await client.writeJournal({ name: "Session 12", folder: "Sessions" });

writeJournalPage(journalIdentifier, pageData: JournalPageData) → Promise<object>

Creates or replaces a page inside a journal entry. Matches an existing page by id or name.

interface JournalPageData {
  id?:    string;
  name?:  string;
  type?:  "text" | "image" | "pdf" | "video"; // default: "text"
  text?: {
    content?:  string;  // raw HTML
    markdown?: string;  // markdown source (when format = 2)
    format?:   1 | 2;  // 1 = HTML (default), 2 = Markdown
  };
  src?:   string;       // media URI for image/pdf/video pages
  title?: { show?: boolean; level?: 1|2|3|4|5|6 };
  flags?: Record<string, unknown>;
}

await client.writeJournalPage("Session 12", {
  name: "Summary",
  type: "text",
  text: { content: "<p>The party entered the dungeon.</p>" },
});

appendJournalPage(journalIdentifier, pageName, html, maxPageBytes?) → Promise<object>

Appends HTML to a page, auto-rotating to a new page (Transcript (2), Transcript (3), …) when the current page exceeds maxPageBytes (default 50000). Creates the first page if none exists.

Ideal for transcription — each utterance is one call, pages never grow unboundedly.

await client.appendJournalPage("Session 12", "Transcript", "<p><b>GM:</b> You enter the tavern.</p>");
await client.appendJournalPage("Session 12", "Transcript", "<p><b>Alice:</b> I look around.</p>");
// rotates to "Transcript (2)" automatically once 50 KB is reached

CLI

A CLI is included for quick scripting via an .env file:

# .env
FOUNDRY_URL=http://localhost:30000
FOUNDRY_USER=<userId>
FOUNDRY_PASS=<password>

npx beavers-client listJournals
npx beavers-client listJournals "My Folder"
npx beavers-client readJournal "My Journal"
npx beavers-client writeJournal '{"name":"My Journal","folder":"My Folder"}'
npx beavers-client writeJournalPage "My Journal" '{"name":"Page 1","text":{"content":"<p>Hi</p>"}}'

License

MIT