Security audit
Security checks for vulnerabilities and agentic risk
This is a coherent plugin quality-review skill, with only minor activation-scope cautions from broad triggers.
This skill is reasonable to install for plugin review workflows. Before installing, be aware it may activate on broad words like review, quality, validation, testing, or architecture, and its intended use involves running local repository commands such as git diff, make test, make lint, make typecheck, and Python validation scripts.
Using 'review' as a trigger creates a shadowing risk with a built-in command of the same name, which can cause the wrong handler to run or make invocation behavior ambiguous. In agent environments, command/skill name collisions are dangerous because they can hijack expected control flow, suppress the intended built-in action, or produce inconsistent results depending on resolution order.
The trigger list includes broad, everyday terms like 'review', 'quality', 'validation', 'testing', and 'architecture', which can cause the skill to activate in many unrelated contexts. In an agent system, unintended activation can lead to incorrect workflow routing, loading unnecessary modules, or causing this skill to influence tasks that were not meant to invoke plugin-review behavior.
No suspicious patterns detected.
