hyperframes

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read and follow directions from project files like DESIGN.md and visual-style.md to define visual identities and constraints. This ingestion of external data creates a potential surface for indirect prompt injection.
  • Ingestion points: Project-specific files including DESIGN.md and visual-style.md.
  • Boundary markers: No specific delimiters or warnings to ignore embedded instructions are mentioned.
  • Capability inventory: File read/write access via platform tools and shell command execution using the hyperframes CLI.
  • Sanitization: No explicit sanitization or validation of the ingested file content is performed before interpolation into the agent's context.
  • [COMMAND_EXECUTION]: The skill defines a workflow involving several shell commands for project initialization, linting, validation, and rendering using the hyperframes utility and local scripts (e.g., animation-map.mjs).
  • [EXTERNAL_DOWNLOADS]: Sample HTML templates utilize the GSAP library fetched from the well-known cdn.jsdelivr.net content delivery network.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 01:25 AM
Security Audit — agent-trust-hub — hyperframes