finetuning

Diagnose NCCL failures and adjacent training-pod failures on HyperPod GPU clusters (EKS or Slurm) — training hangs, AllReduce / collective-op timeouts, EFA or libfabric errors, rendezvous failures, EFA TCP fallback, /dev/shm or memlock issues, NCCL version mismatch across pods, container OOM / exit-137 / OOMKilled, GPU OOM (CUDA out of memory), CrashLoopBackOff / Pending pods, MASTER_ADDR DNS, NetworkPolicy blocking. Not for single-node hardware faults (→ hyperpod-node-debugger § G) or cluster-creation EFA / SSM failures (→ hyperpod-cluster-debugger § A / § F).

Diagnostic-only skill for Slurm scheduler and node-daemon issues on Amazon SageMaker HyperPod Slurm clusters. Scope mirrors the HyperPod troubleshooting guide. Invoke when the user reports a Slurm node stuck in down/drain, "Node unexpectedly rebooted" after auto-repair, slurmd not running, jobs stuck PENDING with REASON=Resources while sinfo shows idle nodes, jobs stuck COMPLETING after node replacement, GRES/GPU counts wrong, scontrol ping failing, slurmctld unresponsive, an Action:Reboot/Replace request that did not trigger HyperPod auto-recovery, or auto-resume not restarting a job. Also triggers on "drain before reboot", "diagnose a Slurm node", "investigate stuck jobs."

Diagnose performance issues on Amazon SageMaker HyperPod clusters — uneven NCCL bandwidth across nodes and poor filesystem throughput. Read-only. Surfaces host-side signals (Xid, ECC, NVLink, EFA reachability, FSx saturation) and routes to the appropriate sibling skill (hyperpod-node-debugger, hyperpod-nccl, hyperpod-version-checker, hyperpod-issue-report) for any remediation. Triggers on uneven NCCL across nodes, straggler node, FSx slow, checkpoint slow, dataloader slow, filesystem bottleneck, FSx throughput, cross-AZ latency, topology mismatch.

Diagnose and remediate per-node issues on a HyperPod cluster (EKS or Slurm) — a specific node is unhealthy, unresponsive, stuck, or needs replacing. Covers on-node EFA, GPU / accelerator hardware (XID, ECC, NVLink, row-remap, DCGM), Slurm node down/drained, disk and memory pressure, per-node lifecycle-script failures, SSM agent, container runtime, kernel panics, pod networking. Read-only. Not for cluster-wide provisioning (→ hyperpod-cluster-debugger), NCCL (→ hyperpod-nccl), or MFU (→ hyperpod-mfu-debugger).

Diagnose and remediate cluster-wide HyperPod (EKS or Slurm) problems — creation / deployment failures (CloudFormation, EFA health check, lifecycle scripts, capacity), EKS access, node replacement, CloudFormation nested-stack errors, post-maintenance rollback state, dangling nodes, autoscaler conflicts. Includes `--validate` pre-flight. Read-only.

Selects a fine-tuning technique (SFT, DPO, RLVR, or RLAIF) for the user's use case and validates it against the selected model's available recipes. Use when the user has decided to finetune and needs to choose a technique, or when the technique needs to be validated against a model. Requires a base model to already be selected (via model-selection skill).

Remote command execution and file transfer on SageMaker HyperPod cluster nodes via AWS Systems Manager (SSM). This is the primary interface for accessing HyperPod nodes — direct SSH is not available. Use when any skill, workflow, or user request needs to execute commands on cluster nodes, upload files to nodes, read/download files from nodes, run diagnostics, install packages, or perform any operation requiring shell access to HyperPod instances. Other HyperPod skills depend on this skill for all node-level operations.

Generate comprehensive issue reports from HyperPod clusters (EKS and Slurm) by collecting diagnostic logs and configurations for troubleshooting and AWS Support cases. Use when users need to collect diagnostics from HyperPod cluster nodes, generate issue reports for AWS Support, investigate node failures or performance problems, document cluster state, or create diagnostic snapshots. Triggers on requests involving issue reports, diagnostic collection, support case preparation, or cluster troubleshooting that requires gathering logs and system information from multiple nodes.

Scan a codebase for deepening opportunities, present them as a visual HTML report, then grill through whichever one you pick.

Diagnosis loop for hard bugs and performance regressions. Use when the user says "diagnose"/"debug this", or reports something broken/throwing/failing/slow.

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes: Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/spec asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use when the user wants to review a branch, a PR, work-in-progress changes, or asks to "review since X".

Investigate a question against high-trust primary sources and capture the findings as a Markdown file in the repo. Use when the user wants a topic researched, docs or API facts gathered, or reading legwork delegated to a background agent.

This skill should be used when the user wants to "write agent code", "build an agent with ADK", "add a tool", "create a callback", "define an agent", "use state management" — in a project that needs ADK (Agent Development Kit) API patterns and code examples. It provides a quick reference for agent types, tool definitions, orchestration patterns, callbacks, state management, the graph Workflow API, and reference recipes to study. Do NOT use for scaffolding (use google-agents-cli-scaffold) or deployment (use google-agents-cli-deploy).

Set up Claude Code hooks to block dangerous git commands (push, reset --hard, clean, branch -D, etc.) before they execute. Use when user wants to prevent destructive git operations, add git safety hooks, or block git push/reset in Claude Code.

创建 lark-cli 的自定义 Skill。当用户需要把飞书 API 操作封装成可复用的 Skill（包装原子 API 或编排多步流程）时使用。

Generate AI music on RunComfy via the `runcomfy` CLI — a smart router across the music-model catalog. Routes to ElevenLabs AI Music Generation (premium 44.1 kHz stereo vocal tracks, 5 s–5 min, $0.0083/s) and ACE Step / ACE Step 1.5 (StepFun-AI open-weights, tag-driven composition, multilingual lyrics, $0.0002–0.0003/s, ~27× cheaper), plus ACE Step audio-inpaint (regenerate a time range inside an existing track) and ACE Step audio-outpaint (extend a track before or after). Picks the right model for the user's actual intent — premium vocal hook, cheap background music library, multilingual pop song, repair a bad chorus, lengthen a 30 s draft into a 2 min cut — and ships each model's documented prompting patterns plus the minimal `runcomfy run` invoke. Triggers on "generate music", "make a song", "AI music", "background music", "instrumental track", "soundtrack", "jingle", "theme music", "royalty-free music", "compose", "music with lyrics", "extend music", "fix this song", "inpaint music", or any explicit ask to generate or edit music.