fish-audio-tts
Pass
Audited by Gen Agent Trust Hub on Nov 4, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill correctly instructs users to store API keys in environment variables (
.env) rather than hardcoding them, following secret management best practices. - [SAFE]: The skill references the official and well-known Fish Audio API endpoint (
api.fish.audio) for legitimate text-to-speech generation functionality. - [INDIRECT_PROMPT_INJECTION]: The skill provides an interface for processing user-supplied text to generate audio narration, which is an inherent surface for indirect prompt injection.
- Ingestion points: The
textinput parameter in theFishAudioTTSandTTSSelectorusage examples withinSKILL.md. - Boundary markers: Absent; there are no specific delimiters or instructions for the agent to ignore potentially malicious commands embedded in the narration text.
- Capability inventory: The skill performs network requests to an external API and writes generated audio data to the local file system via the
output_pathparameter. - Sanitization: No sanitization or input validation logic is described for the text input before it is sent to the TTS service.
Audit Metadata