paraglide-js

Pass

Audited by Gen Agent Trust Hub on Sep 19, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill references official documentation and integration guides from inlang.com and tanstack.com. These are well-known technology services and the official homes for the libraries discussed.
  • [COMMAND_EXECUTION]: The instructions include standard developer workflows using npx @inlang/paraglide-js for project initialization and message compilation. These commands are the documented method for using the library.
  • [INDIRECT_PROMPT_INJECTION]: The skill identifies surfaces where untrusted data (e.g., messages/*.json) is ingested to generate code artifacts in src/paraglide/.
  • Ingestion points: External message files and project configuration (SKILL.md, references/basics.md).
  • Boundary markers: Not explicitly defined; the skill relies on the underlying compiler's logic.
  • Capability inventory: Uses npx to run the compiler which writes to the local file system (references/basics.md).
  • Sanitization: Relies on the standard Paraglide JS compiler to handle message escaping and generation safely.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 19, 2026, 06:43 PM
Security Audit — agent-trust-hub — paraglide-js