seo-drift
Pass
Audited by Gen Agent Trust Hub on Jun 16, 2026
Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes several internal Python scripts (e.g.,
scripts/drift_baseline.py,scripts/drift_compare.py,scripts/drift_report.py) to capture SEO baselines, compare states, and generate HTML reports. - [DATA_EXFILTRATION]: Performs network requests to user-provided URLs to fetch HTML and performance data. The skill documentation indicates it uses a validation mechanism to mitigate SSRF (Server-Side Request Forgery) risks by blocking private IPs and reserved ranges.
- [PROMPT_INJECTION]: The skill processes untrusted external data (HTML tags, headings, and JSON-LD schema) from target URLs. This creates a surface for indirect prompt injection where malicious content on a scanned page could attempt to influence the agent's analysis of SEO changes. The risk is minimized by the skill's focus on structured data extraction and comparison rather than open-ended content generation.
Audit Metadata