voice-audio-engineer

Pass

Audited by Gen Agent Trust Hub on Mar 5, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface
  • Ingestion points: The skill utilizes mcp__firecrawl__firecrawl_search and WebFetch to ingest untrusted data from the internet. It also reads local audio files (via Read) which are processed for transcription.
  • Boundary markers: The skill lacks explicit instructions or delimiters to prevent the agent from obeying malicious commands embedded in external web content or audio transcripts.
  • Capability inventory: The agent has high-privilege tool access, including Bash, Write, Edit, and the ability to create new AI agents via mcp__ElevenLabs__create_agent. These could be abused if an indirect injection occurs.
  • Sanitization: There is no evidence of sanitization, filtering, or validation of data fetched from external sources before it is processed by the agent.
  • [COMMAND_EXECUTION]: Local Command Execution
  • The skill explicitly allows the use of the Bash tool. While intended for audio processing and file management, this provides an execution environment that could be targeted by prompt injection attacks.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 5, 2026, 09:06 PM