fish-audio-tts

Pass

Audited by Gen Agent Trust Hub on Oct 26, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted user-supplied text for conversion into speech via an external API.
  • Ingestion points: The text input field used in TTSSelector().execute() and FishAudioTTS().execute() (SKILL.md).
  • Boundary markers: There are no explicit instructions or delimiters provided to ensure that instructions embedded within the user's text are not interpreted by the agent or the downstream TTS model.
  • Capability inventory: The skill possesses the ability to make network requests to api.fish.audio and perform file write operations to the local filesystem via the output_path parameter.
  • Sanitization: The instructions do not specify any sanitization, filtering, or escaping logic for the text input before it is processed.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 26, 2026, 06:31 PM