content-engine

Pass

Audited by Gen Agent Trust Hub on Jun 16, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill consists of natural language instructions for content strategy and drafting social media posts.
  • [NO_CODE]: No scripts, code files, or executable patterns were detected in the skill definition.
  • [PROMPT_INJECTION]: The skill describes a repurposing flow that processes external assets (articles, videos, memos), which is a data ingestion surface. 1. Ingestion points: source asset (SKILL.md). 2. Boundary markers: Absent. 3. Capability inventory: No file system, network, or code execution tools are requested or used. 4. Sanitization: Not specified. This surface is considered safe because the skill has no dangerous capabilities that could be leveraged by malicious input.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 16, 2026, 11:09 AM
Security Audit — agent-trust-hub — content-engine