skills/whizzzkid/skills/wk-refactor/Gen Agent Trust Hub

wk-refactor

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes Bash, git, and GitHub CLI (gh) tools to analyze pull requests and code repositories. These commands are integral to its refactor validation logic, such as fetching merge bases and generating diffs.
  • [PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface because it ingests and analyzes untrusted data from source code and git diffs (via git diff and git show). Maliciously crafted content within the analyzed codebase (e.g., in comments or removed lines) could theoretically attempt to influence the agent's reasoning during the 'Removed-line audit'.
  • Ingestion points: Git diff output and source files retrieved from the repository in Stages 1, 2, and 3.
  • Boundary markers: Absent. The instructions do not provide specific delimiters or 'ignore embedded instructions' warnings when the agent processes code content.
  • Capability inventory: The agent has access to Bash (for git/gh operations), Write (for report generation), and AskUserQuestion (for interactive auditing).
  • Sanitization: None detected. The code content is processed directly for semantic analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 09:50 PM
Security Audit — agent-trust-hub — wk-refactor