thumb-first-design

Pass

Audited by Gen Agent Trust Hub on Jun 15, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by ingesting untrusted data such as source code, screenshots, and URLs for analysis. Ingestion points: Ingestion occurs in 'workflows/audit.md' Step 1 during the design review process. Boundary markers: No explicit delimiters or instructions for the agent to ignore embedded instructions in the audited content are defined. Capability inventory: The skill utilizes file system writing, browser tools, and shell command execution to generate and display reports. Sanitization: No sanitization of the input data is specified in the workflows.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands like 'open', 'xdg-open', or 'start' to view the generated HTML audit reports in the default browser. This is a functional feature of the skill used to provide the user with a rendered version of the findings.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 15, 2026, 12:14 AM
Security Audit — agent-trust-hub — thumb-first-design