databricks-model-serving
Manage Databricks Model Serving endpoints via CLI. Use when asked to create, configure, query, or manage model serving endpoints for LLM inference, custom models, or external models.
90
88%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Passed
No known issues
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 a well-crafted skill description that concisely covers what the skill does, when to use it, and includes strong trigger terms. It uses proper third-person voice, provides explicit trigger guidance with a 'Use when' clause, and is distinctive enough to avoid conflicts with other skills.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'create, configure, query, or manage model serving endpoints' and specifies the domain clearly as Databricks Model Serving via CLI. | 3 / 3 |
Completeness | Clearly answers both 'what' (manage Databricks Model Serving endpoints via CLI) and 'when' (explicit 'Use when asked to create, configure, query, or manage model serving endpoints for LLM inference, custom models, or external models'). | 3 / 3 |
Trigger Term Quality | Includes strong natural keywords users would say: 'Databricks', 'Model Serving', 'endpoints', 'CLI', 'LLM inference', 'custom models', 'external models', 'create', 'configure', 'query'. Good coverage of terms a user working with Databricks serving would naturally use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with clear niche: Databricks-specific, Model Serving-specific, CLI-specific. The combination of 'Databricks', 'Model Serving', and 'endpoints' makes it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
77%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a strong, actionable skill with excellent executable examples and clear workflow sequencing including validation steps and troubleshooting. Its main weakness is moderate verbosity — some sections repeat guidance (CLI discovery emphasis) and the app integration section adds significant length that could be offloaded. The structure is logical but would benefit from splitting the integration content into a referenced file.
Suggestions
Consolidate the CLI discovery guidance into a single concise callout rather than repeating 'run -h first' in multiple places.
Consider moving the 'Integrate with a Databricks App' section to a separate reference file, keeping only a brief pointer in the main SKILL.md.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | Generally efficient but has some unnecessary explanations (e.g., explaining what Model Serving is, the endpoint types table could be leaner). The CLI discovery section repeats itself — it says 'run -h first' three times in slightly different ways. The endpoint structure ASCII diagram is useful but the bullet explanations below it partially duplicate it. | 2 / 3 |
Actionability | Provides fully executable CLI commands with concrete JSON payloads for create, query, and integration. Commands include profile flags, specific JSON structures, and clear flag usage. The 'Other Commands' table gives exact subcommand names with notes on gotchas (e.g., permissions using ID not name). | 3 / 3 |
Workflow Clarity | Multi-step processes are clearly sequenced: create → poll for READY → query. The app integration section has explicit Step 1/Step 2 with a conditional branch. Validation checkpoints are present (check state.ready, use build-logs for debugging). The troubleshooting table serves as an error recovery guide for common failure modes. | 3 / 3 |
Progressive Disclosure | The skill references a parent 'databricks-core' skill and a 'databricks-apps' skill appropriately, and links to a model-serving guide. However, there are no bundle files provided, and the 'What's Next' app integration section is quite lengthy inline content that could arguably be in a separate reference file. The main content is well-structured with headers but runs long for a single file. | 2 / 3 |
Total | 10 / 12 Passed |
Validation
90%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 10 / 11 Passed
Validation for skill structure
| Criteria | Description | Result |
|---|---|---|
frontmatter_unknown_keys | Unknown frontmatter key(s) found; consider removing or moving to metadata | Warning |
Total | 10 / 11 Passed | |
- Repository
- databricks/devhub
- Commit
6338825
- 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.