skills/erphq/skills/offboarding/Gen Agent Trust Hub

offboarding

Pass

Audited by Gen Agent Trust Hub on Jul 6, 2026

Risk Level: SAFENO_CODE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified in the analysis of this skill. The content is strictly limited to organizational process documentation.
  • [PROMPT_INJECTION]: The skill does not contain any instructions intended to override agent behavior, bypass safety guardrails, or reveal system prompts.
  • [DATA_EXFILTRATION]: While the skill describes processes involving sensitive human resources and payroll data, it lacks any network-capable commands or code that could facilitate unauthorized data exposure or exfiltration.
  • [REMOTE_CODE_EXECUTION]: No external dependencies, package installations, or remote script downloads (e.g., curl/wget piped to a shell) are present in the documentation.
  • [NO_CODE]: This skill consists entirely of markdown-based instructions and metadata. It does not include any accompanying scripts, executables, or dynamic execution patterns.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 6, 2026, 04:43 PM
Security Audit — agent-trust-hub — offboarding