yzr-skill-creator
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the subprocess.run function to execute external binaries including markdownlint, ruff, and the opencode CLI. These operations are limited to the skill's workspace and are used for static analysis and evaluation workflows.
- [DYNAMIC_EXECUTION]: The skill automates the execution of sub-agents using opencode run to verify skill functionality and optimize triggering descriptions. It employs a dedicated 'judge' agent with a restricted permission set that denies all tool access to protect the host environment.
- [INDIRECT_PROMPT_INJECTION]: As a tool that audits and modifies other skills, it reads and parses untrusted markdown and code. This creates an attack surface for indirect prompt injection, although the risk is mitigated by the skill's restricted sub-agent execution model.
Audit Metadata