qzcli
Manage GPU compute jobs on the Qizhi (启智) platform using qzcli — a kubectl-style CLI tool. Use when user says "qzcli", "启智平台", "submit job", "stop job", "查计算组", "avail", "list jobs", "batch submit", or needs to manage distributed training jobs on a Qizhi instance.
84
82%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Quality
Discovery
100%Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.
This is an excellent skill description that clearly identifies the tool (qzcli), the platform (Qizhi/启智), and the specific actions it supports. It includes a comprehensive 'Use when' clause with both English and Chinese trigger terms, making it highly discoverable. The description is concise, specific, and distinctive with minimal risk of conflicting with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple concrete actions: manage GPU compute jobs, submit job, stop job, batch submit, list jobs, check compute groups, and manage distributed training jobs. Also names the specific tool (qzcli) and platform (Qizhi/启智). | 3 / 3 |
Completeness | Clearly answers both 'what' (manage GPU compute jobs on the Qizhi platform using qzcli) and 'when' (explicit 'Use when' clause with a comprehensive list of trigger terms and scenarios). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural trigger terms including both English and Chinese variants: 'qzcli', '启智平台', 'submit job', 'stop job', '查计算组', 'avail', 'list jobs', 'batch submit', and 'distributed training jobs'. These are terms users would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive — targets a specific platform (Qizhi/启智), a specific CLI tool (qzcli), and a specific domain (GPU compute job management). Very unlikely to conflict with other skills due to the niche focus and unique trigger terms. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
64%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid, highly actionable CLI reference skill with excellent concrete examples and good structural organization. Its main weaknesses are length (could benefit from splitting advanced topics into referenced files) and the lack of explicit validation checkpoints in multi-step workflows like batch submission. The troubleshooting table is a nice practical addition.
Suggestions
Add explicit validation checkpoints to batch submission workflow: e.g., '1. Preview with --dry-run, 2. Verify job count and parameters, 3. Only then run without --dry-run'
Consider splitting batch config format, HPC jobs, and troubleshooting into separate referenced files to reduce the main skill's token footprint
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is mostly efficient with concrete commands and tables, but it's quite long (~250 lines) and includes some sections that could be trimmed (e.g., the env-var passthrough section, the HPC/CPU jobs section). Some parameter tables and examples could be more compact, though nothing egregiously explains concepts Claude already knows. | 2 / 3 |
Actionability | Excellent actionability throughout — every section provides copy-paste ready commands with real flags and arguments. The batch config JSON example is complete and executable, parameter tables are specific, and both interactive and non-interactive workflows are covered with concrete examples. | 3 / 3 |
Workflow Clarity | The Quick Start provides a clear 4-step sequence, and job submission has logical flow. However, for batch submission (a potentially destructive batch operation), the --dry-run is mentioned but not positioned as a mandatory validation checkpoint before actual submission. There's no explicit 'validate then proceed' feedback loop for batch or create workflows. | 2 / 3 |
Progressive Disclosure | The content is well-organized with clear section headers and logical grouping, but it's entirely monolithic — all content is inline in a single file with no references to supporting documents. The HPC section, batch config format details, and troubleshooting table could reasonably be split into separate reference files for a skill of this length. | 2 / 3 |
Total | 9 / 12 Passed |
Validation
81%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 9 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
allowed_tools_field | 'allowed-tools' contains unusual tool name(s) | Warning |
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 9 / 11 Passed | |
- Commit
2028ac4
- Reviewed
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.