logic-diff

Pass

Audited by Gen Agent Trust Hub on Jul 5, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: The skill implements a structured analytical process for code comparison without invoking external tools or executing code.- [NO_CODE]: No scripts or binaries are included; the skill relies solely on natural language instructions to guide the agent's reasoning.- [DATA_EXPOSURE]: No hardcoded credentials or sensitive file paths were identified in the skill instructions or guide.- [PROMPT_INJECTION]: While the skill ingests user-provided code snippets, it does so within an analytical context and lacks the system capabilities (such as network access or shell execution) required for an indirect prompt injection to pose a functional security risk.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 5, 2026, 09:37 AM
Security Audit — agent-trust-hub — logic-diff