Scaffold a new HyperBEAM Erlang device with its test file. Use when user says 'create a device', 'new Erlang device', 'scaffold a device', or 'add a device'. Not for building full devices with iteration — use /build-device.
Kubernetes expert specialized in easy-db-lab's K3s cluster, Fabric8 manifest builders, and K8s-based deployments. Use for K8s architecture questions, manifest building, pod debugging, or understanding how easy-db-lab uses Kubernetes. This is a Q&A expert, not an executor.
Tiltfile, Makefile, and Kustomize templates for K8s operators
Diagnose and fix Kubernetes pods, services, networking, storage, and rollout failures with kubectl. Safety-first, read-only by default.
Resolve skill names through layered namespace lookup, then preload the valid skills as visible startup messages.
Process every nested-memory trigger before wiping the trigger set so subsequent iterations only see new paths.
Validate remote numeric override maps defensively so null, non-finite, or negative flag payloads never corrupt runtime thresholds.
Delay surfacing max-output-token failures until the recovery loop completes, then emit sanitized tool_result errors.
Resolve skill names through layered namespace lookup, then preload the valid skills as visible startup messages.
Start a non-blocking, turn-scoped memory prefetch that auto-disposes and only injects results once per turn.
Efficiently bring upstream MatClaw updates into a customized install, with preview, selective cherry-pick, and low token usage.
Modo autónomo nocturno — ejecuta tareas de bajo riesgo en bucle, genera PRs pendientes de revisión humana
Validate remote numeric override maps defensively so null, non-finite, or negative flag payloads never corrupt runtime thresholds.
Delay surfacing max-output-token failures until the recovery loop completes, then emit sanitized tool_result errors.
Start a non-blocking, turn-scoped memory prefetch that auto-disposes and only injects results once per turn.
Process every nested-memory trigger before wiping the trigger set so subsequent iterations only see new paths.
Normalize the main thread to a concrete empty-string key so agent-scoped state maps treat main and forked threads uniformly.
利用LLM Agent自动对Verilog RTL设计进行拆分，按always块生成多个子模块
Normalize the main thread to a concrete empty-string key so agent-scoped state maps treat main and forked threads uniformly.
Cloud migration strategies for AWS, GCP, and Azure. Use for lift-and-shift, replatforming, and cloud-native architecture design.
Triage workflow for dependency vulnerabilities: determine reachability, impact, and safe upgrade/remediation plan.
