dara-denney

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Indirect Prompt Injection] (LOW): The skill is designed to ingest data from external MCP tools, which constitutes an attack surface for indirect prompt injection. \n
  • Ingestion points: Information is retrieved through the mcp__persona-agent__retrieve_mental_models, mcp__persona-agent__retrieve_core_beliefs, and mcp__persona-agent__retrieve_transcripts tools. \n
  • Boundary markers: There are no explicit delimiters or instructions to ignore embedded commands within the retrieved text. \n
  • Capability inventory: The skill only has text synthesis capabilities and does not have access to high-risk tools for command execution, file system access, or network operations. \n
  • Sanitization: The workflow does not specify any sanitization or validation for the content returned by the tools. \n- [Prompt Injection] (SAFE): The skill uses strong directives (e.g., 'CRITICAL', 'MUST') to ensure consistent persona behavior, but these do not attempt to bypass safety filters or extract system information.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:22 PM