ops-inspector

AIOps-style CloudBase inspection skill (v3). Use when users need health checks, log diagnosis, alarm interpretation (CPU alert normal?, peak QPS), metrics via queryEnv(action=metrics), or fault playbooks for 429 / function 404 / ACCESS_TOKEN_INVALID / zero invocations. Triggers on 巡检, 诊断, 告警, 峰值 QPS, 限频, 调用量为 0, troubleshooting.

Quality

91%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Low

Low-risk findings worth noting

Quality

Content

86%Weight 40%Scale 1-5

Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.

A well-structured, highly actionable inspection skill: executable tool calls, a sequenced workflow with checkpoints, and properly split-out references. The main improvement area is conciseness — the AIOps Methodology section and severity table add mild padding over what Claude already knows.

Suggestions

Trim or remove the 'AIOps Methodology' section — its principles (data collection, pattern recognition, baseline comparison) are already implied by the concrete workflow steps.

Make the env-check and CLS-disabled branches explicit feedback loops (e.g. 'if envQuery fails, stop and ask the user to bind an environment') rather than implicit.

Dimension	Reasoning	Score
Conciseness	Efficient throughout — tool maps, query patterns, and the report template are tight and assume Claude's competence — but the 'AIOps Methodology' section restates generic principles Claude already knows and the severity table is basic, leaving minor trimming opportunities; not a 5 because of these light pads.	4 / 5
Actionability	Fully executable guidance: exact MCP calls with concrete 'action=' parameters, specific 'metricName' values, ready-made 'queryString' patterns, and a copy-paste report template covering the common inspection cases.	5 / 5
Workflow Clarity	A clear 7-step full-inspection sequence with checkpoints (env check, CLS-status warning, playbook jumps) and a minimal checklist; not a 5 because some validation branches (e.g. what to do when envQuery fails) are implicit rather than explicit.	4 / 5
Progressive Disclosure	Clean overview body with two well-signaled one-level-deep references ('references/alarm-interpretation.md', 'references/fault-playbooks.md') that exist as real bundle files, and clearly labeled sibling-skill navigation; detailed baselines and playbooks are appropriately split out of the main file.	5 / 5
	Total	18 / 20 Passed

Description

96%Weight 40%Scale 1-5

Based on the skill's description, can an agent find and select it at the right time? Clear, specific descriptions lead to better discovery.

A high-quality description: concrete capabilities, explicit use-when guidance, and rich natural trigger terms in both Chinese and English. Its only soft spot is sharing the CloudBase domain with closely related sibling skills, which leaves a small conflict-risk gap from a 5.

Suggestions

Add a short read-only/diagnostic-vs-implementation disambiguator to the description (e.g. 'read-only inspection; for fixes use the implementation skills') to further reduce overlap with sibling CloudBase skills.

Consider dropping the '(v3)' version marker from the description — version info is time-sensitive and adds little trigger value for a user phrasing a request.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions — 'health checks', 'log diagnosis', 'alarm interpretation (CPU alert normal?, peak QPS)', 'metrics via queryEnv(action=metrics)', and 'fault playbooks for 429 / function 404 / ACCESS_TOKEN_INVALID / zero invocations' — giving comprehensive coverage of the inspection domain.	5 / 5
Completeness	Explicitly answers both: the 'what' ('AIOps-style CloudBase inspection skill') and the 'when' ('Use when users need...' and 'Triggers on...'), with concrete trigger phrases attached.	5 / 5
Trigger Term Quality	Comprehensive natural trigger terms across languages — '巡检, 诊断, 告警, 峰值 QPS, 限频, 调用量为 0, troubleshooting' plus 'health checks' and 'log diagnosis' — covering synonyms and the specific error-code phrases users would actually say.	5 / 5
Distinctiveness Conflict Risk	A clear read-only diagnostic niche with distinct trigger terms (巡检/告警/峰值QPS) and a named tool, but it shares the CloudBase surface with sibling implementation skills, leaving minor overlap risk; not a level-5 because the niche is one of several related CloudBase skills.	4 / 5
	Total	19 / 20 Passed

Validation

93%

Warnings & errors only

Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.

Validation — 15 / 16 Passed

Validation for skill structure

Criteria	Description	Result
frontmatter_unknown_keys	Unknown frontmatter key(s) found; consider removing or moving to metadata	Warning

	Total	15 / 16 Passed

Repository: TencentCloudBase/CloudBase-AI-Toolkit
Path: plugin/cloudbase/skills/ops-inspector/SKILL.md
Commit: f07c33c

Reviewed: 4 days ago

Table of Contents

Discovery Implementation Validation

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.