The Agent Skills Directory

[SAFE]: The skill demonstrates secure practices for managing sensitive information by instructing users to store API keys in environment variables (.env) rather than hardcoding them.
[SAFE]: All network communications are directed to legitimate endpoints belonging to ByteDance/Volcengine (openspeech.bytedance.com), which is a well-known service for the described TTS functionality.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to process arbitrary text input for speech synthesis, representing a potential indirect prompt injection surface.
Ingestion points: The text parameter accepted by the TTSSelector and DoubaoTTS tools as shown in SKILL.md examples.
Boundary markers: The instructions do not specify explicit delimiters or "ignore instructions" prompts for the input text, though this is typical for TTS utilities.
Capability inventory: The underlying tools perform network requests to the TTS provider and write files to the local file system (output_path, metadata_path).
Sanitization: No specific input sanitization or filtering is described in the integration guide.

doubao-tts