webgl-landing-steering

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to process user-provided brand adjectives, conversion goals, and technical constraints to generate design recommendations and code templates. This creates a surface where malicious instructions could be embedded in the user's input to influence the agent's behavior.
  • Ingestion points: The skill takes input for brand signals, device mix, motion tolerance, and production constraints in SKILL.md (Section 1 and Section 6).
  • Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the prompt templates.
  • Capability inventory: No executable tools, scripts, network operations, or file-system write capabilities are detected in this skill.
  • Sanitization: The skill does not implement any sanitization or validation for the user-supplied input strings.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 01:59 PM
Security Audit — agent-trust-hub — webgl-landing-steering