kobe-perspective

Pass

Audited by Gen Agent Trust Hub on Sep 1, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill defines a complex persona that processes user input to generate advice and analysis. This creates a surface for indirect prompt injection, where user-supplied data could contain hidden instructions intended to override the persona or agent behavior.
  • Ingestion points: User queries provided to the agent when the kobe-perspective is active (as demonstrated in examples/demo-conversation.md).
  • Boundary markers: Absent; there are no specific instructions or delimiters used to separate user input from the skill's persona logic or to warn the agent about embedded instructions.
  • Capability inventory: The skill is entirely text-based and does not include any scripts, tools, file-system access, or network capabilities that could be exploited.
  • Sanitization: No input validation or sanitization mechanisms are implemented to filter or escape user-provided content before it is processed by the persona.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 1, 2026, 08:01 AM
Security Audit — agent-trust-hub — kobe-perspective