brand-builder

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • COMMAND_EXECUTION (SAFE): The skill executes local shell commands (git log, cat, jq) to gather context about the project. These commands are read-only and target standard project files (README.md, package.json) to populate the brand profile.
  • INDIRECT_PROMPT_INJECTION (SAFE):
  • Ingestion points: The skill reads external content from README.md, package.json, and git log.
  • Boundary markers: Absent; the content is piped directly into the generation prompt.
  • Capability inventory: The skill has the ability to write a YAML file to the local directory or home folder.
  • Sanitization: None; the LLM processes the raw text to extract brand attributes. Given the restricted scope (generating a brand profile), the risk of significant harm is minimal.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 08:35 PM