skills/kuanghs/agent-skills/xfyun-tts/Gen Agent Trust Hub

xfyun-tts

Warn

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DATA_EXFILTRATION]: In scripts/tts.py, the SimpleWebSocket class explicitly disables SSL certificate verification by setting ctx.verify_mode = ssl.CERT_NONE and ctx.check_hostname = False. This allows an attacker to intercept the communication between the agent and iFlytek's servers, potentially capturing the XFYUN_API_KEY, XFYUN_API_SECRET, and the text content being processed.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection as it ingests and processes untrusted external data.
  • Ingestion points: The scripts/tts.py script reads text from positional arguments, files via the --file flag, and standard input (stdin).
  • Boundary markers: There are no boundary markers or instructions provided to the agent to ignore potentially malicious instructions embedded in the synthesized text.
  • Capability inventory: The script has the capability to perform network requests (WebSocket) and write files to the local system.
  • Sanitization: The script performs Base64 encoding on the input text for transport in the JSON payload, but it does not sanitize or validate the content for instructions.
  • [COMMAND_EXECUTION]: The skill executes network operations to connect to the iFlytek WebSocket API and performs file system writes to save the resulting audio data to paths provided via user arguments.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 13, 2026, 10:01 PM
Security Audit — agent-trust-hub — xfyun-tts