Security audit
Security checks for vulnerabilities and agentic risk
This skill is a coding-agent launcher whose powerful file, shell, and git access is disclosed but too broadly triggered and weakly scoped for automatic use.
Install only if you are comfortable with a skill that may delegate coding tasks to OpenCode with read/write file access, shell commands, and git operations. Use it in trusted project directories, avoid ambiguous prompts, and review spawned-agent actions before allowing changes to important repositories or sensitive files.
The activation text is broad enough to trigger on many ordinary coding requests, which can cause the system to invoke a powerful coding subagent when the user did not explicitly request it. In this skill, that risk is amplified because the subagent is described as having file-system and shell access, so an over-broad trigger can expand the scope of actions taken on the host environment.
The skill explicitly advertises full file-system and shell access but does not pair that capability with clear safety constraints, consent requirements, or warnings about destructive/system-impacting behavior. In the context of an auto-triggerable coding agent, this creates a real risk of unintended file changes, command execution, credential exposure, or broader workstation compromise if the skill is invoked on untrusted or ambiguous tasks.
No suspicious patterns detected.
