storybook-setup

Pass

Audited by Gen Agent Trust Hub on Sep 3, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to run npx storybook skills get setup. This executes code from the Storybook CLI. As the skill author is 'storybookjs', this is considered a vendor-provided resource.
  • [PROMPT_INJECTION]: The skill includes instructions like "Follow the printed Markdown precisely" and "Do not substitute your own plan," which attempt to override the agent's decision-making process in favor of external content.
  • [INDIRECT_PROMPT_INJECTION]: The skill creates an attack surface where instructions generated by an external command (npx storybook skills get setup) are processed as authoritative by the agent.
  • Ingestion points: The output of the npx storybook command.
  • Boundary markers: Absent.
  • Capability inventory: The agent is expected to create or modify files (specifically the Storybook preview file and stories).
  • Sanitization: None mentioned.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 3, 2026, 03:15 PM
Security Audit — agent-trust-hub — storybook-setup