What it is
Claude Code plugin that asks Jev which model is good enough for the current task, then routes work to a cheaper capable option when possible.
What it does: Per-task routing judgments before the coding model runs.
Why it belongs in the showcase: Clear Routing showcase tied to agent-model-router / model-routing patterns.
Caveats: Plugin APIs may change; claimed savings are author-reported.
Evidence
Public README; Claude Code plugin surface.
Caveats
- Claude Code plugin surface may change
- Cost savings are author-reported — measure yourself
Unofficial case page. Stars / list position observed for heat; do not treat heat as a quality SLA. Review the upstream README before you depend on it.