skills/delorenj/skills/cartesia-api/Gen Agent Trust Hub

cartesia-api

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references and fetches documentation, README files, and code examples from official Cartesia GitHub repositories (github.com/cartesia-ai, raw.githubusercontent.com/cartesia-ai) and official documentation domains (docs.cartesia.ai, cartesia.sh). These downloads are associated with the primary service provider.
  • [COMMAND_EXECUTION]: The skill includes instructions for installing official SDKs using standard package managers (pip install cartesia, npm i @cartesia/cartesia-js) and provides a documented curl example for API verification. These are standard developer onboarding steps.
  • [INDIRECT_PROMPT_INJECTION]: As the skill facilitates speech-to-text (STT) transcription via the Ink model, it creates a surface where external audio data is converted to text and ingested into the agent context. The instructions emphasize verbatim reading of transcripts, and while explicit prompt boundary markers for the transcribed text are not defined in the skill itself, the ingestion of untrusted external content represents a known surface for indirect injection that requires typical caution in downstream application logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:55 AM
Security Audit — agent-trust-hub — cartesia-api