seo-article-writer

Pass

Audited by Gen Agent Trust Hub on May 5, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection from untrusted sources.
  • Ingestion points: In Step 2, the skill reads content from a 'brief file or pasted content' provided by the user.
  • Boundary markers: There are no explicit boundary markers or instructions to the LLM to ignore potentially malicious instructions embedded within the SEO brief.
  • Capability inventory: The skill has significant capabilities including writing files (Write tool in Step 6), fetching sensitive marketing references (get_marketing_reference), and invoking other skills (Skill: tigerdata-marketing-skills:content-reviewer in Step 10).
  • Sanitization: There is no evidence of sanitization or validation of the input brief before it is processed by the agent.
  • [DATA_EXFILTRATION]: The skill accesses a 'No Fly List' (Step 0) which contains sensitive business information about customers. While the skill includes explicit instructions to protect this data, a successful prompt injection could theoretically attempt to exfiltrate this list via web search or inclusion in the generated article artifact.
Audit Metadata
Risk Level
SAFE
Analyzed
May 5, 2026, 07:20 AM