barry-onboarding

Pass

Audited by Gen Agent Trust Hub on Jun 14, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: Indirect Prompt Injection Surface. The skill reads the content of ~/.barry-english/profile.md to determine the user's progress and select the appropriate onboarding path. This ingestion of local data creates a potential attack surface if the file is modified by an untrusted external process.\n
  • Ingestion points: Step 0 reads state from ~/.barry-english/profile.md.\n
  • Boundary markers: The skill uses structural markdown anchors (e.g., ## Section [1-4]) but lacks explicit delimiters or instructions to ignore embedded commands within those sections.\n
  • Capability inventory: The skill can invoke other functional skills (barry-profile, barry-assessment, barry-solution) and informs the user about existing executable scripts like Barry-Quiz.command.\n
  • Sanitization: There is no evidence of sanitization or content filtering for the data read from the profile file before it is interpreted by the agent.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 14, 2026, 02:43 PM
Security Audit — agent-trust-hub — barry-onboarding