cms-article-publish

Pass

Audited by Gen Agent Trust Hub on Sep 9, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input including article titles, HTML paragraphs, and embed URLs that are passed to CMS management tools.
  • Ingestion points: User-provided input for 'Title' and 'Body' paragraphs in SKILL.md.
  • Boundary markers: Step 2 of the workflow requires a full preview and explicit user confirmation.
  • Capability inventory: Includes tools for creating and updating articles and paragraphs (cms_create_article, cms_create_article_paragraph).
  • Sanitization: No explicit sanitization or instructions to disregard directives within the ingested content are present.
  • [SAFE]: The skill implements a human-in-the-loop validation process (dry-run) that prevents the automated execution of potentially malicious content hidden in user data. It also includes guardrails for input types, timezone confirmation, and scheduling logic.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 9, 2026, 02:57 PM
Security Audit — agent-trust-hub — cms-article-publish