banner-design

Pass

Audited by Gen Agent Trust Hub on Aug 30, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes local Python and Node.js scripts located in the .claude/skills/ directory to facilitate image generation and capturing screenshots of design work. These operations are restricted to specific, non-malicious paths related to the skill's creative purpose.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external data from docs/brand-guidelines.md and user-provided requirements through AskUserQuestion.
  • Ingestion points: User requirements collected via AskUserQuestion and brand guidelines file located at docs/brand-guidelines.md.
  • Boundary markers: None explicitly defined for untrusted text ingestion.
  • Capability inventory: Execution of Python/Node scripts via subprocess calls and browser control through chrome-devtools.
  • Sanitization: None detected for brand guidelines or user-input interpolation.
  • [SAFE]: All external references and design research operations (e.g., Pinterest) are standard industry practices and do not involve sensitive data exfiltration or unauthorized access.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 30, 2026, 01:11 PM
Security Audit — agent-trust-hub — banner-design