tutorializer

Command-line access to your tutorial-video data on app.tutorializer.com: projects, tutorial definitions, rendered videos, the synthesized voiceover speeches behind them, and per-project pronunciation rules.

Install

npm install -g tutorializer   # global install
npx tutorializer --help       # or run without installing

Requires Node.js 18 or newer.

Quick start

tutorializer login            # browser or API-key login (interactive chooser)
tutorializer projects list    # your projects with their ids
tutorializer videos list --projectId PROJECT_ID

Found 2 project(s):
1. Example App (64a1f2c9e4b0a1b2c3d4e5f6)
2. Example Docs (64a1f2c9e4b0a1b2c3d4e5f7)

Signing up

No account yet? Create one from the terminal — the generated password prints exactly once, and the session is stored so every other command works immediately:

tutorializer signup --email [email protected] --json

Authentication

Two ways in:

• Browser session — tutorializer login --browser starts a temporary localhost server, opens your browser to app.tutorializer.com, and receives the session tokens on the redirect back. Tokens are stored in ~/.tutorializer/ and reused by every later command.
• API key — tutorializer login --with-key prompts (masked — never a command argument) for an API key secret created in the app, verifies it with one authenticated call, and stores it. Alternatively export TUTORIALIZER_API_KEY=<key secret> and skip the login step entirely — the key is read from the environment at request time. New keys default to read-only scopes; widen a key in the app before using write commands.

A stored browser session takes precedence over a stored key, which takes precedence over TUTORIALIZER_API_KEY. tutorializer logout clears the stored session and any stored key (an environment key stays in effect).

Without credentials, a command run in an interactive terminal starts the browser login on its own; run non-interactively (no TTY) it fails immediately with exit code 1 and login instructions on stderr — it never opens a browser headlessly.

Commands

Session

tutorializer login              # choose browser or API key interactively
tutorializer login --browser    # log in via browser; tokens land in ~/.tutorializer/
tutorializer login --with-key   # paste an API key secret (masked prompt)
tutorializer logout             # clear the stored session and any stored key

Projects

tutorializer projects list                    # list projects (-n/--limit, default 10)
tutorializer projects get [projectId]         # raw JSON for one project, or the first N
tutorializer projects read <projectId>        # formatted view: name, domain, timestamps
tutorializer projects create <name> --domain example.com   # --domain is optional

Pronunciation rules

Project-wide dictionary entries applied when new voiceover audio is generated:

tutorializer projects pronunciation list --projectId <id>
tutorializer projects pronunciation add <word> <phonemes> --projectId <id> --alphabet ipa
tutorializer projects pronunciation remove <word> --projectId <id>

add adds or replaces the rule for that exact word. --alphabet accepts ipa (default) or cmu. Changing a rule does not regenerate existing audio — use speeches set-pronunciation to regenerate one existing segment.

Speeches

tutorializer speeches list --projectId <id> --language en -n 20   # default 10
tutorializer speeches get [speechId]      # raw JSON for one speech, or a filtered list
tutorializer speeches read <speechId>     # formatted view: text, language, url, duration
tutorializer speeches set-pronunciation <speechId> "<phonetic text>"
tutorializer speeches delete <speechId>
tutorializer speeches purge --contains <substring> --projectId <id> [--language en] [--dry-run] [--yes]

set-pronunciation stores the pronunciation hint and triggers a server-side regeneration of that speech's audio. delete permanently removes the speech record plus its audio and alignment files. purge deletes every speech in a project whose text contains a substring (case-insensitive); it prints the match list first, previews with --dry-run, and prompts for confirmation unless --yes is passed.

Videos

tutorializer videos list --projectId <id> --tutorialId <id> -n 20   # default 10
tutorializer videos get [videoId]       # raw JSON for one video, or a filtered list
tutorializer videos read <videoId>      # formatted view: status, times, url
tutorializer videos delete <videoId>    # permanently remove one rendered output

Tutorials

tutorializer tutorials list --projectId <id> -n 20    # default 10
tutorializer tutorials get [tutorialId]   # raw JSON for one tutorial, or a filtered list
tutorializer tutorials read <tutorialId>  # formatted view

Skills

tutorializer skills list               # names and descriptions of the bundled agent guides
tutorializer skills get tutorializer   # print a bundled SKILL.md to stdout

Schema

tutorializer schema                    # the whole command tree as JSON
tutorializer schema speeches list      # just that command's options and arguments

JSON output

Every data and mutation subcommand accepts --json: real, parseable JSON on stdout with no color. Lists and reads print the records; mutations print a small result object such as { "ok": true, "id": "…" }. Failures print a single-line JSON error object on stderr and exit with code 1:

tutorializer projects list --json
tutorializer speeches delete SPEECH_ID --json   # {"ok": true, "id": "…"}

Configuration

A tutorializer.json in the working directory (or any parent) supplies the default --projectId:

{ "projectId": "64a1f2c9e4b0a1b2c3d4e5f6" }

Explicit --projectId flags win over the config file. TUTORIALIZER_API_URL overrides the API endpoint (default https://api.tutorializer.com) and TUTORIALIZER_APP_URL the browser-login origin (default https://app.tutorializer.com) — only needed against a non-production deployment. TUTORIALIZER_API_KEY supplies an API key secret for headless use (see Authentication).

Usage with AI agents

Install the Tutorializer agent skills — tutorializer (this CLI) and tutorial-library-audit (the MCP-based audit workflow) — for Claude Code, Cursor, Codex, and any other agent that supports the Skills standard:

npx skills add tutorializer/skills

The same guides ship inside the npm package, version-matched to the installed CLI:

tutorializer skills list               # what is bundled
tutorializer skills get tutorializer   # the CLI guide matching this version

Or paste this into your AGENTS.md / CLAUDE.md:

## Tutorial videos

Use the `tutorializer` CLI for tutorial-video data: projects, tutorial
definitions, rendered videos, voiceover speeches, and pronunciation rules.
Run `npx tutorializer skills get tutorializer` for the full guide, and
`tutorializer --help` for the command reference. Log in once with
`tutorializer login`.

Prefer a connector? The Tutorializer MCP server at https://mcp.tutorializer.com/mcp exposes the same data as tools for Claude, ChatGPT, and any MCP-capable host — see tutorializer.com/developers.

License

Apache-2.0