text-to-speech
Pass
Audited by Gen Agent Trust Hub on Aug 1, 2026
Risk Level: SAFEDATA_EXFILTRATIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [SAFE]: The skill follows security best practices for AI agent tools, including proper credential management and restricted network communication.\n- [DATA_EXFILTRATION]: The skill utilizes the HEYGEN_API_KEY environment variable to authenticate requests sent to api.heygen.com. This is the intended behavior for the service and does not constitute unauthorized exfiltration.\n- [EXTERNAL_DOWNLOADS]: The skill documentation and code examples reference fetching configuration and audio previews from official HeyGen domains (api.heygen.com, resource.heygen.ai). These references target a well-known service and are considered safe.\n- [PROMPT_INJECTION]: The skill defines a text-to-speech interface that accepts external strings for processing.\n
- Ingestion points: Untrusted data enters the agent context via the text parameter in the mcp__heygen__text_to_speech tool and related functions in SKILL.md.\n
- Boundary markers: No explicit delimiters or instructions to ignore embedded commands are included in the prompt templates.\n
- Capability inventory: The skill possesses network capabilities to interact with external APIs via fetch and requests.\n
- Sanitization: There is no evidence of input filtering or escaping before data is sent to the remote API. \n As the output is restricted to speech audio generation, this surface represents minimal risk to the agent's core execution environment.
Audit Metadata