BC huawei-cloud-cloudrobo-train
Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAIN_FROM_SCRATCH) and finetune (MODEL_TUNING) tasks with FFT/SFT/LORA/QLORA/DEEPSPEED methods; manage the full task lifecycle (create/read/update/delete/stop/restart/resume/draft); save and resubmit draft configs; count tasks by status; monitor execution stages, resource usage, training logs, signed URLs, and events; diagnose failures. SimRL tasks (simulation reinforcement learning) are managed via the same CLI with a --sim-rl flag and cover the same CRUD + lifecycle + monitoring surface (except resume, which is train-only). Triggers include: model training, fine-tuning, pretraining, training task, training stages, resource usage, training logs, training events, draft task, restart training, stop training, resume training, task stats, simulation reinforcement learning, SimRL, 仿真强化学习, 模型训练, 模型微调, 训练任务, 训练阶段, 资源使用, 训练日志, 训练事件, 草稿任务, 重启训练, 克隆训练, 停止训练, 续训训练, 任务统计.
Manage CloudRobo model training tasks and simulation reinforcement learning (SimRL) tasks — create pretrain (TRAINFROMSCRATCH) and finetune (MODELTUNING)…
As a process C 62/100 · Has gaps — weak spots: result and completion, when it triggers, execution cost
How to improve
- Say in the description WHEN to use the skill ("use when…", example requests): that is the agent's main cue.
- The SKILL.md body is over 5,000 tokens: move reference detail into references/ and load it when needed.
- Your own cases (evals/evals.json, 4–6 real requests with expected answers): the full check would then run those instead of a model-drafted suite.
- A spec.yaml with trigger phrases and assertions — a behaviour contract for CI; `skilltest init` writes a template.
Guard findings · 0
✓ No critical or high findings
Files scanned: 12. Evidence is masked. Grey chips explain why severity was lowered.
Against the Agent Skills spec
- warning
description-no-whendescription does not say WHEN to use the skill (no "use when") - warning
body-longSKILL.md body ≈ 12548 tokens (recommended < 5000); move details to references/
Process rating: all ten parameters 62/100
- 20When it triggers. No condition that starts the skill
- 40Result and completion. Does not say what the result is
- 40Execution cost. Instruction body is 12548 tokens: crowds the task out of the window
- 60Tools and files. Uses tools (bash, web, python) that frontmatter does not declare
- 60Steps. 174 steps, 5 vague phrases
- 70Inputs and preconditions. Inputs and preconditions are listed
- 100Failures and branches. 10 branches, has a failure section
- 100Consistency. Name and required fields are in place
- 100Running it twice. Mutating operations check current state
- 100Progress reporting. Reports progress
- medium Safety rules and hard prohibitions inside a skill: they belong in the system prompt, here they protect nothing
- low The response is described with custom markup (27 tags): a typed call is more reliable
Everything here is measured from the skill text rather than judged by a model, so the numbers are checkable. A parameter weighs more when it is a more common reason for the process to stall.
Quality signals
- +5Description has no quoted example phrases that should trigger the skill
- +4Description does not say when NOT to use the skill (false activations)
- +3Description length 983: 120–800 characters recommended
- +3Output format is not stated: the model decides each time
- +1No license
- +2Single-language instructions
- +4Structure: 41 headings
- +3Step-by-step instructions: 174 items
- +4Has examples (17 code blocks)
- +4Reference files are cited in the instructions (8 of 8)
- +3All 1 scripts are documented
Quality base 70; lint remarks subtract, signals add up to 100. Result: 66.