Ukhradviser

Pass

Audited by Gen Agent Trust Hub on Jun 19, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill instructions are focused entirely on providing professional HR guidance and legal compliance information. No override markers, bypass instructions, or malicious obfuscation were found.
  • [DATA_EXPOSURE]: The skill references a local file path (${HOME}/.claude/skills/UkHrAdviser/CLAUDE.md) to load its extended knowledge base. This is a standard mechanism for skill configuration on the target platform and does not involve accessing sensitive user credentials or system files.
  • [COMMAND_EXECUTION]: No unauthorized command execution or subprocess spawning patterns were identified. The suggested web searches for legislation and ACAS guidance are part of the intended professional functionality.
  • [PROMPT_INJECTION]: The skill maintains a structured workflow (Understand, Check, Assess, Advise) which serves as a natural guardrail against simple direct injection attempts by focusing the agent on specific procedural outputs.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 19, 2026, 06:14 PM
Security Audit — agent-trust-hub — Ukhradviser