vertex-engine-inspector
Inspect and validate Vertex AI Agent Engine deployments including Code Execution Sandbox, Memory Bank, A2A protocol compliance, and security posture. Generates production readiness scores. Use when asked to inspect, validate, or audit an Agent Engine deployment. Trigger with "inspect agent engine", "validate agent engine deployment", "check agent engine config", "audit agent engine security", "agent engine readiness check", "vertex engine health", or "reasoning engine status".
75
71%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Advisory
Suggest reviewing before use
Optimize this skill with Tessl
npx tessl skill review --optimize ./plugins/ai-ml/jeremy-vertex-engine/skills/vertex-engine-inspector/SKILL.mdQuality
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 a strong skill description that clearly defines its scope around Vertex AI Agent Engine deployment inspection and validation. It lists specific capabilities, provides explicit trigger guidance with multiple natural variations, and occupies a distinct niche that minimizes conflict risk. The use of third person voice and concrete terminology makes it well-suited for skill selection.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: inspect and validate deployments, covers specific components (Code Execution Sandbox, Memory Bank, A2A protocol compliance, security posture), and generates production readiness scores. | 3 / 3 |
Completeness | Clearly answers both 'what' (inspect/validate Vertex AI Agent Engine deployments across multiple dimensions, generate readiness scores) and 'when' (explicit 'Use when...' clause with detailed trigger phrases). | 3 / 3 |
Trigger Term Quality | Provides extensive natural trigger terms including variations like 'inspect agent engine', 'validate agent engine deployment', 'check agent engine config', 'audit agent engine security', 'agent engine readiness check', 'vertex engine health', and 'reasoning engine status'. These cover multiple natural phrasings a user might use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly specific niche targeting Vertex AI Agent Engine deployments specifically, with distinct domain-specific triggers like 'reasoning engine status' and 'A2A protocol compliance' that are unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
42%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
The skill is well-structured with good progressive disclosure and a clear workflow sequence, but critically lacks actionable, executable content. For a technical inspection skill, the absence of any concrete code snippets (SDK calls, curl commands, monitoring queries, scoring calculations) means Claude would need to invent all implementation details. The scenarios section describes use cases narratively rather than showing concrete input/output examples.
Suggestions
Add executable Python SDK code snippets for key operations: retrieving agent metadata (`client.agent_engines.get()`), listing agents, and querying configuration fields—these are the core actions Claude needs to perform.
Include concrete curl commands for A2A protocol testing (e.g., `curl -s https://<endpoint>/.well-known/agent-card`) and actual Cloud Monitoring API filter expressions for querying error rates and latency.
Provide the weighted scoring formula explicitly (e.g., category weights, threshold definitions, how 0-100% maps to pass/fail) rather than deferring all scoring logic to reference files.
Replace the narrative scenario descriptions with at least one concrete example showing actual input (agent ID, project) and expected output (YAML report snippet with scores).
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill includes some unnecessary verbosity—the overview restates the description, the prerequisites section explains what tools are for, and the examples section describes scenarios narratively rather than showing concrete inputs/outputs. However, it's not egregiously padded and the error table is efficient. | 2 / 3 |
Actionability | Despite describing a complex multi-step inspection process, the skill provides zero executable code, no concrete SDK calls beyond a method name hint, no actual monitoring queries, no curl commands for A2A testing, and no scoring formula. The instructions read as abstract descriptions rather than copy-paste-ready guidance. | 1 / 3 |
Workflow Clarity | The 10-step numbered sequence provides a clear order of operations, but there are no validation checkpoints, no feedback loops for when checks fail mid-inspection, and no explicit decision points (e.g., what to do if a category scores below threshold). The actual inspection logic is deferred to reference files. | 2 / 3 |
Progressive Disclosure | The skill effectively uses one-level-deep references to separate detailed content: inspection-workflow.md, inspection-categories.md, example-inspection-report.md, and errors.md are all clearly signaled with descriptive context. The main file serves as a well-organized overview with appropriate section structure. | 3 / 3 |
Total | 8 / 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
70e9fa4
- 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.