website-to-hyperframes
Pass
Audited by Gen Agent Trust Hub on May 17, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill makes extensive use of the
npx hyperframescommand suite for various stages of the video production pipeline, including site capture, linting, validation, and rendering. These commands interact with the local file system and network. - [INDIRECT_PROMPT_INJECTION]: In Step 1, the skill captures content from user-provided URLs and instructs the agent to read and summarize the extracted text, headings, and CTA buttons. This creates a surface for indirect prompt injection if the target website contains malicious instructions designed to subvert the agent's logic during subsequent steps like scriptwriting or storyboard generation.
- Ingestion points:
references/step-1-capture.mdinstructs the agent to readcapture/extracted/visible-text.txtandasset-descriptions.mdderived from the external URL. - Boundary markers: Absent. The instructions do not provide delimiters or warnings to ignore instructions embedded within the captured website content.
- Capability inventory: The skill possesses network access (
hyperframes capture), file-writing capabilities (creating project files), and execution capabilities (rendering and validating compositions). - Sanitization: Absent. The agent is directed to process and incorporate extracted text directly into artifacts.
- [DYNAMIC_EXECUTION]: The skill involves the dynamic generation of HTML/JavaScript compositions based on data extracted from external websites. These compositions are later executed in a browser environment via
npx hyperframes validateandpreview. While this is the core functionality of the tool, assembling code from untrusted external data sources presents a potential vector for cross-site scripting (XSS) or logic manipulation within the rendering environment. - [DATA_EXPOSURE]: The skill references the use of
.envfiles for managingGEMINI_API_KEYorGOOGLE_API_KEY. While storing secrets in environment files is a standard security practice, it confirms that the skill environment manages sensitive credentials.
Audit Metadata