greenfield-clean-breaks

Pass

Audited by Gen Agent Trust Hub on Jun 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [SAFE]: The skill's instructions and reference templates are focused exclusively on code refactoring and architectural simplification when compatibility pressure is removed.\n- [COMMAND_EXECUTION]: Uses the rg (ripgrep) command-line tool to identify code patterns and call sites as part of the refactoring review loop. This is a standard development operation and is used within the intended scope of the skill.\n- [PROMPT_INJECTION]: The skill presents an indirect prompt injection surface as it processes external source code provided by the user. No malicious exploits are present in the skill itself.\n
  • Ingestion points: Source code files specified in the /goal target (e.g., SKILL.md).\n
  • Boundary markers: Absent; the instructions do not define specific delimiters for processed code content.\n
  • Capability inventory: Includes file reading via rg and file modification/deletion using the agent's standard tooling.\n
  • Sanitization: Absent; the skill does not provide methods for escaping or validating the contents of the target files.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 24, 2026, 06:19 PM
Security Audit — agent-trust-hub — greenfield-clean-breaks