skills/hk-hub/agentskills/shader-dev/Gen Agent Trust Hub

shader-dev

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFENO_CODEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill consists of markdown documentation and code templates for computer graphics programming. All provided logic is related to mathematical rendering techniques like Signed Distance Fields (SDF), Ray Marching, and Fluid Simulation.
  • [NO_CODE]: The skill provides static templates and instructions for an AI agent to generate code for a user. It does not include scripts intended to be executed on the host system by the agent during its own runtime.
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a surface for processing user-supplied requests through the $ARGUMENTS variable to generate procedural graphics code.
  • Ingestion points: User-provided descriptions of desired visual effects in SKILL.md.
  • Boundary markers: None identified in the provided templates.
  • Capability inventory: File creation/writing (HTML, JS, and GLSL) based on templates.
  • Sanitization: Not applicable as the skill provides static educational content.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 03:43 AM
Security Audit — agent-trust-hub — shader-dev