← ClaudeAtlas

panellisted

Cross-vendor entry point + adversarial panel engine. `/xm:panel <verb>` routes multi-model work to the matching consumer in --cross-vendor mode (review→x-review, plan(brainstorm)/debate/council→x-op, solve→x-solver, eval→x-eval, consensus→x-build, fan-out→x-agent); `/xm:panel <target>` runs the panel engine itself (N model CLIs refute each other → consensus verdict); bare `/xm:panel` is an interactive picker; cross/detect/doctor/preflight/types/models are engine utilities (preflight = live model check before a run). Use for "panel review", "다른 모델들로 같이 리뷰", "여러 LLM으로 적대 리뷰", "다중모델로 토론/문제해결/평가", "panel 돌리기 전에 모델/프로바이더 상태 점검", or /xm:panel.
x-mesh/xm · ★ 0 · AI & Automation · score 70
Install: claude install-skill x-mesh/xm
# x-panel — Cross-Model Adversarial Review Panel ## Overview `x-panel` is the **multi-model entry point**. It has two jobs: 1. **Router** — `/xm:panel <verb>` (review/debate/council/solve/eval/consensus/fan-out) delegates to the matching consumer (x-review/x-op/x-solver/x-eval/x-build/x-agent) in `--cross-vendor` mode. The domain logic stays in the consumer; panel just picks the door. This is why "do it with several models" has ONE obvious entry point instead of remembering each plugin's flag. 2. **Engine** — `/xm:panel <target>` runs the native panel: N model CLIs review the same target in one round by default, and a verdict separates **consensus** (how many models agreed — confidence) from **diversity** (what only one model saw). The orchestrator is a tool-neutral CLI, so the "leader" is not a fixed model. Different models have different blind spots — in dogfooding, codex missed a perf issue claude/agy caught, and cursor missed a SQL injection. That diversity is the whole point of both jobs. For reviews, these jobs are alternatives, not consecutive stages. `/xm:panel review` routes once to x-review, which owns target selection, lenses, severity, lifecycle, verdict, and convergence; panel only replaces x-review Phase 3 as its execution backend. Never run the native panel afterward as a second review. A bare `xm panel <target>` is an ad-hoc consensus tool and does not create an x-review lifecycle/verdict artifact. ## When to Use - "여러 모델로 같이 리뷰", "다중모