storybook-mcp-integration

Pass

Audited by Gen Agent Trust Hub on Mar 30, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill implements an automated generate-test-heal cycle that relies on ingesting external data from the Storybook MCP server to drive file system modifications.\n
  • Ingestion points: Documentation, component properties, and test results retrieved via list-all-documentation, get-documentation, and run-story-tests.\n
  • Boundary markers: There are no explicit instructions or delimiters used to ensure the agent ignores potential instructions embedded within the component documentation or test error messages.\n
  • Capability inventory: The skill leverages Write and Edit file system tools alongside the ability to execute story tests.\n
  • Sanitization: No sanitization or validation logic is defined to prevent malicious content in stories or documentation from influencing the agent during the self-healing process.\n- [EXTERNAL_DOWNLOADS]: The skill facilitates the installation of the @storybook/addon-mcp and @storybook/addon-vitest packages from the NPM registry and references Chromatic for remote Storybook access.\n- [COMMAND_EXECUTION]: The documentation includes npx commands for upgrading Storybook, adding addons, and registering the MCP server with the agent environment.
Audit Metadata
Risk Level
SAFE
Analyzed
Mar 30, 2026, 03:24 AM