hyperframes-core

Pass

Audited by Gen Agent Trust Hub on Jun 27, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill adheres to established security practices for its domain, utilizing the official hyperframes-cli for project management and trusted CDNs for essential animation and styling libraries.
  • [PROMPT_INJECTION]: The skill documents an orchestration pattern that represents a surface for indirect prompt injection through the processing of user-controlled project files.
  • Ingestion points: The agent reads project plans from STORYBOARD.md and SCRIPT.md to coordinate video generation.
  • Boundary markers: The sub-agent dispatch instructions call for copying context verbatim into worker prompts without explicit delimiters.
  • Capability inventory: The agent utilizes task delegation tools (e.g., Agent tool, spawn_agent) and shell-based validation utilities.
  • Sanitization: No specific sanitization or validation of the plan data is mentioned before it is used to influence sub-agent behavior.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 27, 2026, 11:29 PM
Security Audit — agent-trust-hub — hyperframes-core