hyperframes-general-video
Pass
Audited by Gen Agent Trust Hub on Sep 12, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill uses
npxto interact with the HyperFrames registry for skill updates and component discovery. It also retrieves assets from Figma when a user provides afigma.comURL. - [COMMAND_EXECUTION]: The workflow involves executing shell commands to initialize projects (
npx hyperframes init), check authentication status (npx hyperframes auth status), and run utility scripts for preference recording and packet generation. - [INDIRECT_PROMPT_INJECTION]: Project files like
BRIEF.mdandSTORYBOARD.mdare ingested to drive the creation process. While the skill utilizes dedicated adapters to handle external data (such as SVG sanitization for Figma exports), the integration of these inputs into agent prompts represents a standard risk surface for indirect instructions. - [DYNAMIC_EXECUTION]: The skill dynamically generates role payloads and task packets for sub-agents at runtime. The
scripts/frame-packets.mjsscript assembles these by concatenating framework core instructions with scene-specific data and blueprint definitions.
Audit Metadata