seed-component-map

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates exclusively as a discovery tool for local repository structures. It uses built-in modules (node:fs and node:path) to read directory contents and file metadata. It does not perform any network operations, write to the filesystem, or execute external scripts.
  • [INDIRECT_PROMPT_INJECTION]: The tool ingests untrusted data by reading the content of source files within the repository to extract export patterns and registry configuration.
  • Ingestion points: The mapSeedComponent function in scripts/component-map.ts reads file content using readFile to build an export graph and check registry manifests.
  • Boundary markers: The skill outputs a structured JSON object to stdout, separating the found paths into logical buckets (rootage, headless, implementations, etc.).
  • Capability inventory: The skill's capabilities are restricted to read-only access of specific subdirectories within the repository root. It does not possess write or execution capabilities.
  • Sanitization: Component name inputs are strictly validated for length (max 120 chars) and character set (alphanumeric, spaces, underscores, and hyphens). File content is processed using fixed regular expressions for pattern matching rather than dynamic evaluation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 05:48 AM
Security Audit — agent-trust-hub — seed-component-map