seo-email

Pass

Audited by Gen Agent Trust Hub on Jul 9, 2026

Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [PROMPT_INJECTION]: The skill processes untrusted data fetched from an external SEO API (DataForSEO), which reflects strings from third-party websites, creating an attack surface for indirect prompt injection.
  • Ingestion points: Ingests audit findings and keyword data from ~/.claude/skills/seo/output/<domain>-audit.json.
  • Boundary markers: None. The instructions do not use delimiters or provide warnings to ignore embedded instructions within the ingested audit data.
  • Capability inventory: The skill has file read/write access and executes local scripts (preflight.sh and keyword_research.py).
  • Sanitization: None. Untrusted audit data is interpolated directly into the outreach email draft.
  • [COMMAND_EXECUTION]: Executes local shell and Python scripts for environment initialization and credential verification.
  • Evidence: Instructions require executing ~/.claude/skills/seo/scripts/preflight.sh and keyword_research.py during the preflight phase.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 9, 2026, 11:11 AM
Security Audit — agent-trust-hub — seo-email