Cast

Multilingual TTS via Typecast CLI with emotion control. Plays audio aloud or saves to file.

Security Scan

The skill's requirements and instructions line up with a Typecast CLI TTS integration; nothing requested appears disproportionate to that purpose.

Purpose & Capability

Name/description (Typecast TTS) match the declared binary ('cast') and the single required credential (TYPECAST_API_KEY). The brew install of a 'cast' CLI is coherent with the stated purpose.

Instruction Scope

SKILL.md only shows using the 'cast' CLI, reading user-provided text (echo, cat script.txt), listing/setting voices and config, and saving/playing audio. It references TYPECAST_* env vars that correspond to flags the CLI accepts. There are no instructions to read unrelated system files or to exfiltrate arbitrary data.

Install Mechanism

Installation is via a Homebrew formula from the tap 'neosapience/tap', which is a third-party tap (not the default homebrew/core). This is a common and reasonable install path for vendor-provided CLIs, but you should review the formula before installing because third-party taps run code during install and place binaries on disk.

Credentials

Only TYPECAST_API_KEY is required as a credential and is declared as the primaryEnv. The SKILL.md documents additional optional TYPECAST_* env vars that map to CLI flags; no unrelated secrets or multiple unrelated credentials are requested.

Persistence & Privilege

The skill does not request 'always: true' or other elevated persistence. It's instruction-only and relies on an external binary; it doesn't instruct modifying other skills or system-wide agent settings.

Assessment

This skill appears to do what it says: run the Typecast 'cast' CLI using a Typecast API key. Before installing, verify the Homebrew formula (neosapience/tap/cast) to ensure it comes from the official vendor and doesn't run unexpected install-time code. Treat TYPECAST_API_KEY as a secret: prefer supplying it as an environment variable or a dedicated API key with limited scope; revoke it if you see suspicious activity. If you have strong privacy requirements, confirm what text/audio is sent to Typecast and how long they retain data. Finally, review where the 'cast' CLI stores credentials (the SKILL.md mentions 'cast login <api_key>') so you know whether an API key is stored on disk and where.

cast — Typecast TTS CLI

Quick start

cast "Hello!"                                # play immediately
cast "Hello" --out hello.mp3 --format mp3   # save to file
cast "I can't believe it!" --emotion smart  # AI emotion inference

Voices

cast voices list                            # list all voices
cast voices list --gender female
cast voices list --model ssfm-v30
cast voices get <voice_id>                  # voice details
cast config set voice-id <tc_xxx>           # set default voice

Emotions

• --emotion smart — AI infers emotion from text (ssfm-v30 only)
• --emotion preset --emotion-preset <preset> — explicit preset

preset	model
normal / happy / sad / angry	ssfm-v21, ssfm-v30
whisper / toneup / tonedown	ssfm-v30 only

cast "I'm so happy!" --emotion preset --emotion-preset happy --emotion-intensity 1.5
cast "I just got promoted!" --emotion smart \
  --prev-text "I worked so hard." --next-text "Let's celebrate!"

Key flags

flag	default	description
--voice-id	—	voice ID
--model	ssfm-v30	ssfm-v30 (quality) / ssfm-v21 (low latency)
--tempo	1.0	speed (0.5–2.0)
--pitch	0	pitch (–12 ~ +12)
--volume	100	volume (0–200)
--out	—	output file path
--format	wav	wav / mp3
--seed	—	reproducible output

Recipes

echo "System ready." | cast
cast "$(cat script.txt)"
cast "Hello!" --out /tmp/hi.mp3 --format mp3
cast "Hello" --tempo 1.2 --pitch 2 --emotion preset --emotion-preset happy

Auth / config

cast login <api_key>     # save API key
cast config list         # show current config
cast config set model ssfm-v21

Env vars (TYPECAST_ prefix):

Variable	Flag
TYPECAST_API_KEY	--api-key
TYPECAST_VOICE_ID	--voice-id
TYPECAST_MODEL	--model
TYPECAST_LANGUAGE	--language
TYPECAST_EMOTION	--emotion
TYPECAST_EMOTION_PRESET	--emotion-preset
TYPECAST_EMOTION_INTENSITY	--emotion-intensity
TYPECAST_FORMAT	--format
TYPECAST_VOLUME	--volume
TYPECAST_PITCH	--pitch
TYPECAST_TEMPO	--tempo