seo-content
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill invokes local Python scripts (
word_count.py,link_audit.py,tech_audit.py) from thescripts/directory to perform quality and technical audits on content drafts. - [EXTERNAL_DOWNLOADS]: The
scripts/link_audit.pyscript performs network HEAD requests using theurllib.requestlibrary to verify the status of external URLs provided within content drafts. - [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from external sources like competitor websites and community forums to generate research briefs and content.
- Ingestion points: Competitor pages, Reddit threads, and user-provided draft files.
- Boundary markers: The skill employs a multi-step workflow with defined templates (Brief, Outline, Draft) to compartmentalize information.
- Capability inventory: The skill executes local Python scripts and manages project files in the
.seo/directory. - Sanitization: Audit scripts utilize
argparsefor input handling, but no specific sanitization is performed on external article text before ingestion.
Audit Metadata