wk-workstyle-testing

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No security issues detected. The skill provides clear instructions for the agent to evaluate test quality using benign tools like Read, Glob, and Grep.
  • [DATA_EXPOSURE]: No hardcoded secrets, sensitive file paths, or credentials were found. The skill operates exclusively on project code and test files.
  • [REMOTE_CODE_EXECUTION]: The skill does not include any commands for downloading external scripts, installing packages, or executing remote code.
  • [INDIRECT_PROMPT_INJECTION]: While the skill ingests untrusted code and test data during the review process (SKILL.md), it lacks the high-risk capabilities (such as network access or arbitrary command execution) required to facilitate an exploit chain. The risk is considered low.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 09:51 PM
Security Audit — agent-trust-hub — wk-workstyle-testing