temple-generator

HIGH W007: Insecure credential handling detected in skill instructions.

• Insecure credential handling detected (high risk: 0.90). The skill reads user notes and inlines extracted node content (descriptions, exemplars, and TEMPLE_DATA JSON which can be embedded into HTML), which could cause the LLM to reproduce any secret strings present in the vault verbatim in its outputs.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.70). The template HTML loads and executes remote JavaScript at runtime via the importmap (e.g., https://cdn.jsdelivr.net/npm/three@0.163.0/build/three.module.js and its examples path), which is a required runtime dependency that fetches and runs external code in the renderer.