internal-links
Pass
Audited by Gen Agent Trust Hub on Sep 25, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: No security issues detected. The skill is a standard SEO instructional guide focusing on internal link structure, link equity, and anchor text optimization.
- [SAFE]: The skill does not contain any executable scripts, shell commands, or obfuscated code. It lacks network access and does not attempt to read sensitive system files.
- [INDIRECT_PROMPT_INJECTION]: The skill instructs the agent to read
contextus.mdfor project context. While this is an external data ingestion point, the skill's capabilities are limited to generating SEO recommendations, presenting a negligible risk of malicious instruction execution.
Audit Metadata