compose-page-blocks

Pass

Audited by Gen Agent Trust Hub on Jul 26, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill operates as a deterministic transformation tool. It implements a strict input/output contract where source HTML must be pre-sanitized by an upstream process (html-sanitize.ts) that removes scripts, iframes, and event handlers.
  • [SAFE]: The instructions explicitly forbid the reintroduction of dangerous elements like <script>, <iframe>, <form>, or raw CSS styles inside the generated block markup.
  • [SAFE]: The skill includes a 'no-hallucination' policy enforced by an external verifier (output-verify.ts), ensuring the agent does not invent or modify text from the source document, which prevents the injection of unauthorized content.
  • [SAFE]: The skill limits operations to an allow-list of standard WordPress core blocks and uses abstract design tokens (slugs) from a design-foundation.json file instead of hardcoded hex values, preventing unauthorized styling or layout escapes.
  • [SAFE]: The disable-model-invocation: true flag in the frontmatter indicates that this skill likely provides instructions for a deterministic process or highly constrained generation rather than open-ended model usage.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 26, 2026, 03:02 AM
Security Audit — agent-trust-hub — compose-page-blocks