Skill

jev-use

Use before ANY step that is a decision rather than writing: labelling, filtering or triaging MANY items; checking whether a command, test or build succeeded; picking the next action from options you can list; judging safety, quality or severity; or risk-checking each command of a checklist or pipeline before it runs. Route by where the facts already are: facts already in your context go to the jev_judge tool with ALL questions batched into ONE call; items sitting in a file or in tool output go through the `jev-use judge` CLI from a script, so that data never enters the conversation; a safety decision that blocks every tool call belongs in the `jev-use hook gate` PreToolUse hook, not in a call you make by hand. Take any verdict back with escalate: true. Never for steps that must produce new text or code, or judgments whose options you cannot enumerate.

Bundled with jev-use. Its trust score, security findings and other files are on the plugin page.

npx [email protected] add shitianfang/jev-use