gsp-visuals

Pass

Audited by Gen Agent Trust Hub on May 15, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface through its ingestion of workspace files to gather design context.
  • Ingestion points: Project documentation (e.g., BRIEF.md, .yml configurations) are read in SKILL.md and various domain-specific scripts.
  • Boundary markers: No explicit delimiters or instructions to ignore nested commands are provided.
  • Capability inventory: The skill has access to Write, Glob, Grep, and WebSearch tools.
  • Sanitization: No content validation or sanitization is performed on ingested data. This surface is considered safe as it is intrinsic to the skill's primary design role and operates within local project boundaries.
  • [SAFE]: No indicators of malicious behavior, credential harvesting, obfuscation, or unauthorized remote code execution were found in any of the skill's files.
  • [SAFE]: Tool usage is strictly limited to relevant file operations and search, consistent with the stated design purpose.
Audit Metadata
Risk Level
SAFE
Analyzed
May 15, 2026, 01:19 AM
Security Audit — agent-trust-hub — gsp-visuals