who-am-i-writing-for

Pass

Audited by Gen Agent Trust Hub on Nov 6, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill instructions involve processing untrusted data from external sources, specifically user comments, replies, and messages.
  • Ingestion points: Step 1 and Step 4 direct the agent to analyze the exact wording of comments and messages to extract questions and vocabulary.
  • Boundary markers: The instructions do not define specific delimiters or guardrails to prevent the agent from potentially following instructions hidden within that user-generated content.
  • Capability inventory: The skill references data retrieval tools (get_analytics, list_posts, get_voice_insights) but does not include capabilities for code execution, file system modification, or non-platform network requests.
  • Sanitization: There are no explicit instructions to sanitize or filter the content of processed messages.
  • [SAFE]: The skill methodology focuses on evidence-based analysis and explicitly discourages the invention of fictional data. The integration with the VoiceMoat service (voicemoat.com) is documented neutrally as a way to enhance analytics gathering.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 6, 2026, 07:23 AM