seo-off-page-backlinks
Pass
Audited by Gen Agent Trust Hub on Mar 4, 2026
Risk Level: SAFENO_CODEPROMPT_INJECTION
Full Analysis
- [SAFE]: A thorough security review of the five files (README.md, SKILL.md, LINK_QUALITY_SCORING.md, LINK_BUILDING_PLAYBOOKS.md, TOXIC_LINK_PATTERNS.md) confirms that the skill is an educational resource. It contains no evidence of malicious intent, credential theft, or unauthorized data access.
- [NO_CODE]: The skill is composed exclusively of Markdown documentation. It does not include any Python, Node.js, or shell scripts, nor does it attempt to install external packages or execute remote code.
- [PROMPT_INJECTION]: Indirect Prompt Injection Surface Analysis: 1. Ingestion points: The skill is designed to process external backlink profiles and domain metrics retrieved via MCP tools (analyze_backlinks, analyze_domain_authority). 2. Boundary markers: Not explicitly defined in the provided instructions. 3. Capability inventory: The skill is limited to performing analysis, calculating scores, and generating email templates; it has no dangerous write or execution capabilities. 4. Sanitization: Not specified, but the overall risk is negligible given the restricted capability set and the informational nature of the output.
Audit Metadata