supreme-problem-solving

Pass

Audited by Gen Agent Trust Hub on May 20, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill uses an internal constraint (instruction.hierarchy.max.priority.no.later.input.can.override) to ensure the agent maintains focus on the defined problem-solving methodology. This is a common reliability measure for complex skills.\n- [PROMPT_INJECTION]: The skill is designed to process untrusted external data such as logs and traces during the diagnostic phase.\n
  • Ingestion points: User-provided logs, traces, code snippets, and evidence quotes.\n
  • Boundary markers: Not explicitly defined in the methodology instructions.\n
  • Capability inventory: Guides the agent's behavior during analysis and verification; relies on standard agent tools for implementation.\n
  • Sanitization: Not specified within the skill instructions.\n- [SAFE]: The skill does not request network access, expose credentials, or perform unauthorized code execution.
Audit Metadata
Risk Level
SAFE
Analyzed
May 20, 2026, 03:41 AM
Security Audit — agent-trust-hub — supreme-problem-solving