sveltia-cms

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The initialization script scripts/init-sveltia.sh ingests user-supplied repository names, branch names, and directory paths without sanitization to generate project configuration files.
  • Ingestion points: Interactive read commands within scripts/init-sveltia.sh (e.g., REPO, BRANCH, ADMIN_DIR).
  • Boundary markers: Absent; user-provided strings are interpolated directly into YAML and HTML templates.
  • Capability inventory: The script performs directory creation (mkdir) and file writing (cat >) on the local file system.
  • Sanitization: No validation, escaping, or path-traversal checks are performed on the user-provided inputs.
  • [DYNAMIC_EXECUTION]: The scripts/init-sveltia.sh script dynamically generates an index.html file that includes a script tag pointing to a remote JavaScript module from a CDN.
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads external resources required for the CMS functionality.
  • Fetches the @sveltia/cms library from unpkg.com, a well-known CDN, via HTML templates.
  • Clones the official Sveltia CMS authentication worker from https://github.com/sveltia/sveltia-cms-auth using standard Git commands.
  • These resources are provided by the official CMS vendor or well-known development services.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 06:55 AM
Security Audit — agent-trust-hub — sveltia-cms