context-ramp

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (medium risk: 0.30). The required runtime workflow for narration (render-narration-local.mjs → POST /v1/audio/speech to a loopback NeMo-Speech.cpp server) ingests free text from an outsider-controlled --input file (“narration text file”) and sends it to the TTS server for audio generation.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's runtime scripts explicitly git-clone and build the NVIDIA NeMo runtime (https://github.com/NVIDIA/NeMo-Speech.cpp.git) and call Hugging Face downloads / hf_hub_download for model repos (nvidia/magpie_tts_multilingual_357m, nvidia/nemo-nano-codec-22khz-1.89kbps-21.5fps, hexgrad/Kokoro-82M), which are fetched during runtime and are required to execute the local TTS code and model loading.