databricks-model-serving

Deploy and query Databricks Model Serving endpoints. Use when (1) deploying MLflow models or AI agents to endpoints, (2) creating ChatAgent/ResponsesAgent agents, (3) integrating UC Functions or Vector Search tools, (4) querying deployed endpoints, (5) checking endpoint status. Covers classical ML models, custom pyfunc, and GenAI agents.

Quality

—

Does it follow best practices?

Impact

—

No eval scenarios have been run

Securityby

Passed

No known issues

Quality

Content

65%

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

The body is highly actionable with concrete executable examples, but it suffers from redundant duplicate querying blocks, a large inline endpoint reference table, a missing deploy-status validation checkpoint in the workflow, and nine referenced bundle files that do not exist.

Suggestions

Add an explicit validation checkpoint between deploy and query (e.g., Step 6.5: poll 'manage_serving_endpoint(action="get")' until state is READY before querying), rather than only mentioning polling in the issues table.

Remove the duplicated querying examples in 'Common Workflows' (lines 248-270), which repeat the manage_serving_endpoint query blocks already shown under 'MCP Tools'.

Provide the missing referenced files (1-classical-ml.md through 9-package-requirements.md) or move the large foundation-model endpoint table into one of them so the signaled navigation resolves to real bundle content.

Dimension	Reasoning	Score
Conciseness	Mostly efficient tables and code, but the 'Common Workflows' section duplicates the querying examples already shown (lines 204-226 and 248-270 are near-identical) and the 28-row foundation-model endpoint table is heavy inline reference data that could live in a reference file.	2 / 3
Actionability	Fully executable, copy-paste ready guidance — exact pip versions ('mlflow==3.6.0', 'langgraph==0.3.4'), concrete MCP tool calls with parameters, exact endpoint names, and complete code blocks match the anchor for specific executable examples.	3 / 3
Workflow Clarity	The 7-step Quick Start is clearly sequenced, but the deploy (Step 6) to query (Step 7) transition omits an explicit validation checkpoint to poll for READY state; polling is only mentioned later in the Common Issues table, so per the batch/async guideline workflow clarity is capped at 2.	2 / 3
Progressive Disclosure	The Reference Files table is well-signaled with a 'When to Read' column at one level deep, but all nine referenced bundle files (1-classical-ml.md through 9-package-requirements.md) are missing and no references/ directory exists, so the navigation points to non-existent files and the heavy inline endpoint table should be split out.	2 / 3
	Total	9 / 12 Passed

Description

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.

The description is strong: it states concrete actions, includes explicit numbered 'Use when' triggers, and carves out a distinct niche for Databricks Model Serving. Third-person voice is used throughout. It is a model example of a well-formed skill description.

Dimension	Reasoning	Score
Specificity	Lists multiple concrete actions in third person — 'deploying MLflow models or AI agents to endpoints', 'creating ChatAgent/ResponsesAgent agents', 'integrating UC Functions or Vector Search tools', 'querying deployed endpoints', 'checking endpoint status' — matching the anchor for multiple specific concrete actions.	3 / 3
Completeness	Explicitly answers both 'what' ('Deploy and query Databricks Model Serving endpoints') and 'when' via the numbered 'Use when (1)...(5)...' clause, matching the anchor for clear what-and-when with explicit triggers.	3 / 3
Trigger Term Quality	Good coverage of natural terms users would say — 'deploy', 'query', 'MLflow models', 'AI agents', 'endpoints', 'UC Functions', 'Vector Search', 'endpoint status' — with only minor jargon (ResponsesAgent, pyfunc) preventing a clear margin above anchor 3.	3 / 3
Distinctiveness Conflict Risk	Has a clear niche (Databricks Model Serving) with distinct triggers tied to MLflow/agents/UC Functions/Vector Search; while related skills exist, the description's triggers are specific to model serving and unlikely to fire for unrelated skills.	3 / 3
	Total	12 / 12 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
relative_links	Relative link issues: 15 missing, 5 suspicious	Warning

	Total	15 / 16 Passed

Repository: databricks-solutions/ai-dev-kit
Commit: 1c43c21

Reviewed: 3 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.