Generate Protoss-style (StarCraft) voice effects using SoX and FFmpeg.

Apply Protoss-style (StarCraft) psionic effects to ANY audio file. Use as a post-processing layer for TTS or user recordings.

Security Scan

The skill's code, runtime instructions, and install steps are coherent with its stated purpose (audio post-processing using ffmpeg and sox) and do not request unrelated privileges or network access.

Purpose & Capability

Name/description claim audio post-processing; required binaries (ffmpeg, sox), brew install of ffmpeg/sox, and the included protoss_fx.py all directly implement that. Nothing requested is disproportionate to transforming audio files.

Instruction Scope

SKILL.md instructs the agent to generate/obtain input audio, run the provided script, and deliver the processed file. The script only reads the provided input file, creates local temporary WAVs in the same directory, and writes a single output file — no unrelated file reads, remote endpoints, or broad data collection are present.

Install Mechanism

Install spec uses Homebrew to install ffmpeg and sox (well-known package source). No arbitrary downloads or extract steps that would write unknown code to disk beyond standard packages.

Credentials

The skill requires no environment variables, secrets, or external credentials. The claimed capabilities do not need additional access, and SKILL.md doesn't reference other env vars or tokens.

Persistence & Privilege

always is false and the skill does not attempt to modify agent configuration or persist credentials. It creates temporary files in the working directory and cleans them up; no elevated or permanent presence is requested.

Assessment

This skill appears to only perform local audio processing with ffmpeg and sox. Before installing: (1) confirm you trust the skill owner/source since package provenance is 'unknown'; (2) install ffmpeg/sox from your system package manager (brew) rather than copying binaries from untrusted sites; (3) avoid running the script on untrusted/malicious media files (specially crafted media can sometimes exploit codecs); and (4) review the included protoss_fx.py (it's short and uses subprocess.run with argument lists, not shell=True) if you want extra assurance.

Protoss Voice Effect

Applies a "Khala" psionic transformation chain to audio files using the V9 "Massive Void" engine. Modular Design: This skill does NOT generate speech. It transforms existing audio.

Usage

Run the script on any input audio file (wav, mp3, ogg, etc):

python3 skills/protoss-voice/protoss_fx.py <path_to_audio_file>

Output: Creates a new file with suffix _psionic.wav in the same directory.

Agent Protocol & Behavior

When acting as a Protoss persona (e.g., Selendis, Artanis, Zeratul, etc), the agent should:

1. Generate/Record Base Audio:
   • Use kokoro-tts (or any other TTS skill) to generate the raw speech.
   • OR accept a user-provided recording.
2. Process (The "Black Box"):
   • Execute protoss_fx.py on the raw file.
   • Do not narrate this step unless debugging.
3. Deliver Final Artifact:
   • Send the processed file (_psionic) using the message tool.
   • Clean up raw/intermediate files silently if they are temporary.

Integration Example (Kokoro)

# 1. Generate (Raw)
python3 skills/kokoro-tts/speak.py "En Taro Adun." -o raw.wav -v ef_dora

# 2. Transform (Psionic)
python3 skills/protoss-voice/protoss_fx.py raw.wav
# Output: raw_psionic.wav

# 3. Optimize for Transport (Telegram OGG)
ffmpeg -i raw_psionic.wav -c:a libopus -b:a 64k -vbr on final.ogg -y

# 4. Send
message(action="send", media="final.ogg", asVoice=true)

Requirements

Requires ffmpeg and sox (Sound eXchange).

brew install ffmpeg sox