The Agent Skills Directory

[SAFE]: The skill communicates exclusively with verified HeyGen infrastructure, including api.heygen.com, upload.heygen.com, and files.heygen.ai. No unauthorized or suspicious third-party domains were identified.
[SAFE]: Credential management follows security best practices. All API requests use the HEYGEN_API_KEY environment variable through standard X-Api-Key headers, with no hardcoded secrets present in the instructions or reference files.
[SAFE]: The provided Python and TypeScript code examples are limited to standard API wrappers for video orchestration and asset management. There is no evidence of subprocess spawning, persistence mechanisms, or unauthorized file system operations.
[INDIRECT_PROMPT_INJECTION]: The skill processes external assets via user-provided URLs and populates templates using dynamic variables. While this presents a potential surface for indirect injection or SSRF at the service level, the risk is inherent to the primary functionality and is mitigated by the documentation's focus on structured data and validation.
[METADATA_POISONING]: An automated malware reputation flag was raised for SKILL.md. Manual analysis confirms this is a false positive, as the file contains only descriptive Markdown and standard tool configurations without any executable or obfuscated content.

avatar-video