The Agent Skills Directory

[DYNAMIC_EXECUTION]: The main script (scripts/run.py) employs os.execv to restart the process within a dedicated virtual environment (.runners-venv). This is a legitimate development pattern used to ensure the tool runs with the correct Python interpreter and necessary dependencies.\n- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted text data through the --prompt and --prompt-file parameters to generate audio.\n
Ingestion points: User-provided script text via CLI arguments, local files, or standard input (specified in SKILL.md).\n
Boundary markers: None explicitly defined in the orchestration instructions to distinguish user text from system prompts.\n
Capability inventory: The skill has access to Bash, Write, and Read tools, primarily for script execution and outputting audio files.\n
Sanitization: No explicit sanitization or filtering logic is present in the provided entry point script; however, the untrusted data is passed as a payload to external TTS APIs (OpenAI/ElevenLabs) rather than being executed as local code.

voiceover-maker