chanjing-tts
Fail
Audited by Gen Agent Trust Hub on Mar 14, 2026
Risk Level: HIGHCREDENTIALS_UNSAFEPROMPT_INJECTION
Full Analysis
- [CREDENTIALS_UNSAFE]: The skill includes hardcoded authentication credentials for the Chanjing TTS API within the
Obtain AccessTokensection ofSKILL.md.\n - Evidence: The JSON request body specifies a literal
app_id("84042cb5") andsecret_key("10cd5091fe6042dfb91ba01816a991e0").\n- [PROMPT_INJECTION]: The skill has an indirect prompt injection surface due to its handling of user-provided text for speech synthesis.\n - Ingestion points: The
textfield in theCreate Speech APIcall (SKILL.md).\n - Boundary markers: Absent. No delimiters or instructions are provided to help the agent distinguish between text to be synthesized and potential malicious instructions.\n
- Capability inventory: The skill performs outbound network requests to
open-api.chanjing.cc(SKILL.md).\n - Sanitization: Absent. The instructions do not define any validation or escaping for the input text before transmission to the external API.
Recommendations
- AI detected serious security threats
Audit Metadata