listenhub-voice

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from multiple sources, which is then processed by the agent and sent to an external API. This creates a potential surface for indirect prompt injection attacks where malicious instructions could be embedded in the provided text or metadata of external assets.
  • Ingestion points: Processes user-supplied synthesis text (up to 1400 characters), public URLs for audio voice cloning, and URLs or Base64 data for images.
  • Boundary markers: No explicit delimiters or instructions are provided to the model to ignore embedded commands within the processed script text or external data.
  • Capability inventory: The skill uses curl to perform network operations (POSTing to the API and downloading results) and performs file-write operations to the local filesystem when saving generated audio.
  • Sanitization: The skill employs jq with the --arg flag to safely interpolate user text into JSON payloads, providing protection against basic shell command injection when constructing API requests.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 08:20 PM
Security Audit — agent-trust-hub — listenhub-voice