search-design-system

Pass

Audited by Gen Agent Trust Hub on Aug 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes a local Node.js script scripts/cli.mjs to perform design system searches and artifact retrieval. It uses a structured command-line interface and parses JSON-based search requests to avoid direct shell interpolation.
  • [EXTERNAL_DOWNLOADS]: Fetches design system manifests, indexes, and documentation artifacts (such as DESIGN.md) from the author's official GitHub Pages domain (lintendo.github.io). All downloads are subject to origin validation and mandatory integrity verification via SHA-256 hashing. The fetch mechanism includes size limits and safe redirect handling to prevent resource exhaustion and server-side request forgery.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests external documentation content (DESIGN.md) that could potentially contain instructions aimed at influencing the agent's behavior.
  • Ingestion points: Content is retrieved from remote URLs or local paths during the fetch workflow.
  • Boundary markers: The skill does not explicitly define delimiters when processing retrieved markdown content.
  • Capability inventory: The skill has the capability to perform network requests and manage a local object cache, but it does not have the ability to execute arbitrary shell commands from the downloaded content.
  • Sanitization: While the skill provides robust validation for the location and integrity of the source files, the textual content of the design documentation is processed without specific prompt-injection filtering.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 28, 2026, 10:08 PM
Security Audit — agent-trust-hub — search-design-system