ai-component-metadata

Pass

Audited by Gen Agent Trust Hub on Jun 12, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill provides a local Python utility, scripts/generate_metadata.py, designed to extract component structure and generate documentation metadata. The script performs standard file read/write operations and uses regular expression parsing without network access, external dependencies, or unsafe dynamic evaluation primitives.
  • [SAFE]: The documentation and integration guides (e.g., Figma MCP usage and AI-driven testing) describe legitimate development workflows and architectural patterns for improving AI component recognition.
  • [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection by processing external component files to generate metadata that influences AI behavior.
  • Ingestion points: scripts/generate_metadata.py reads content from user-provided component files in the local filesystem.
  • Boundary markers: The skill does not currently specify boundary markers or instructions to isolate extracted metadata content from the agent's instructions.
  • Capability inventory: The skill utilizes file system access (read/write) through its provided Python script and interacts with design tools such as Figma.
  • Sanitization: There is no explicit sanitization or validation of the text extracted from component files during the metadata generation process.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 12, 2026, 10:42 PM
Security Audit — agent-trust-hub — ai-component-metadata