head-of-design

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection. It is designed to ingest and analyze untrusted external data from the repository, such as marketing pages, documentation, and Storybook files.
  • Ingestion points: Files located in docs/, components, and partner-facing UIs.
  • Boundary markers: No specific delimiters or instructions to ignore embedded commands are present in the audit workflow.
  • Capability inventory: The skill can execute shell commands (git log), write files to the design/ directory, and delegate tasks to the /cto agent, potentially propagating malicious instructions.
  • Sanitization: There is no mention of sanitizing or validating the content of the files before processing.
  • [COMMAND_EXECUTION]: The skill executes local shell commands to gather context.
  • Evidence: git log --oneline -10 -- <surface paths> is used in Step 1 to review recent changes. This is a standard operation for repository analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 04:50 PM
Security Audit — agent-trust-hub — head-of-design