offer-letter

Pass

Audited by Gen Agent Trust Hub on Aug 6, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [PROMPT_INJECTION]: The skill uses natural language instructions to guide document drafting. No patterns of instruction override, jailbreak attempts, or system prompt extraction were found.
  • [DATA_EXFILTRATION]: No network operations or sensitive file access patterns are present. The skill only processes text provided within the user's brief in the local context.
  • [REMOTE_CODE_EXECUTION]: The skill does not include any scripts, package dependencies, or remote content downloads.
  • [COMMAND_EXECUTION]: No shell commands, system calls, or privilege escalation patterns were detected.
  • [NO_CODE]: The skill is comprised entirely of instructions in markdown format and lacks any associated executable logic, scripts, or tool configurations.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 6, 2026, 05:31 AM
Security Audit — agent-trust-hub — offer-letter