Personal Operating Manual

Pass

Audited by Gen Agent Trust Hub on Feb 16, 2026

Risk Level: LOWNO_CODE
Full Analysis
  • [PROMPT_INJECTION] (SAFE): No instructions found that attempt to override agent behavior or bypass safety filters.
  • [DATA_EXFILTRATION] (SAFE): No commands for accessing sensitive files or making network requests were detected.
  • [REMOTE_CODE_EXECUTION] (SAFE): No external dependencies, package installations, or remote script executions are present.
  • [COMMAND_EXECUTION] (SAFE): The skill contains no shell commands or subprocess calls.
  • [OBFUSCATION] (SAFE): No hidden or encoded content (Base64, zero-width characters, etc.) was found.
  • [INDIRECT_PROMPT_INJECTION] (SAFE): The skill does not ingest untrusted external data, eliminating the risk of indirect injection.
Audit Metadata
Risk Level
LOW
Analyzed
Feb 16, 2026, 04:37 AM