hyperframes-best-practices
Pass
Audited by Gen Agent Trust Hub on May 15, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No malicious patterns, obfuscation, or unauthorized data exfiltration were detected. The skill's design, instructions, and utility scripts are consistent with standard professional development and video production workflows.
- [COMMAND_EXECUTION]: The skill and its documentation reference the use of local development tools such as
node,npx, andffmpeg. These are essential for the HyperFrames rendering engine and the provided animation mapping and contrast reporting scripts. - [EXTERNAL_DOWNLOADS]: The skill facilitates the fetching of Hebrew fonts from Google Fonts and suggests the use of well-known external APIs (e.g., ElevenLabs, OpenAI, Google Cloud) for Hebrew text-to-speech, as the built-in TTS engine does not currently support the language. These are legitimate external dependencies required for the skill's purpose.
- [PROMPT_INJECTION]: The skill contains structural instructions like
<HARD-GATE>to enforce design constraints on the agent. These are not malicious bypass attempts but rather guardrails to ensure output quality and consistency with the project's visual identity.
Audit Metadata