Name: migrate-from-model-serving
Rating: 65.60000000000001 (1 reviews)
Author: databricks

migrate-from-model-serving

Migrate an MLflow ResponsesAgent from Databricks Model Serving to Databricks Apps. Use when: (1) User wants to migrate from Model Serving to Apps, (2) User has a ResponsesAgent with predict()/predict_stream() methods, (3) User wants to convert to @invoke/@stream decorators.

Quality

78%

Does it follow best practices?

Run evals on this skill

Adds up to 20 points to the overall score

View guide

Securityby

Passed

No findings from the security scan

Fix and improve this skill with Tessl

tessl review fix ./agent-openai-agents-sdk/.claude/skills/migrate-from-model-serving/SKILL.md

The canonical home for this skill is migrate-from-model-serving in databricks/app-templates

Quality

Content

70%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.

The body is a thorough, well-sequenced migration runbook with strong validation checkpoints, but it is verbose for its length and keeps all reference material inline rather than splitting it across bundle files.

Suggestions

Trim explanatory prose Claude already knows (async concurrency rationale, the await/async-for basics in 'Async Patterns') and de-duplicate repeated reminders like 'uv.lock must be committed' and the --profile note.

Extract the 'Reference: Common Migration Patterns', 'Reference: App File Structure', and 'Troubleshooting' sections into separate files under references/ and link to them one level deep, reducing the monolithic SKILL.md.

Where the core agent.py transformation uses `...`/TODO placeholders, provide at least one complete end-to-end worked example (sync and async) so the migration code is fully copy-paste executable.

Dimension	Reasoning	Score
Conciseness	The ~960-line body is mostly actionable, but padded prose (explaining async concurrency benefits, re-illustrating basic await/async-for mechanics) and repeated reminders ('uv.lock must be committed', '--profile') add tokens Claude does not need.	3 / 5
Actionability	Concrete, copy-paste-ready bash commands and code span every step (download, setup, test, deploy) with a resource mapping table, but core agent transformation examples retain `...`/TODO placeholders that prevent full executability.	4 / 5
Workflow Clarity	A clearly sequenced six-step process with task tracking, explicit validation checkpoints (the 5.5 pre-deploy checklist, `bundle validate`), and troubleshooting feedback loops for error recovery.	5 / 5
Progressive Disclosure	No bundle reference files exist and the entire guide is a single monolithic document, with reference-style sections (file structure, migration patterns, troubleshooting) inlined that would benefit from being split into one-level-deep referenced files.	3 / 5
	Total	15 / 20 Passed

Description

87%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.

The description is specific, third-person, and cleanly answers both what the skill does and when to use it with concrete trigger clauses. Its only weakness is limited synonym coverage in trigger phrasing.

Dimension	Reasoning	Score
Specificity	Names the precise domain (MLflow ResponsesAgent, Model Serving to Apps) and concrete actions (migrate the agent, convert to @invoke/@stream decorators), but the narrow niche yields only ~2 distinct actions, leaving minor coverage gaps versus a fully comprehensive list.	4 / 5
Completeness	Explicitly states both what it does ('Migrate an MLflow ResponsesAgent from Databricks Model Serving to Databricks Apps') and when to use it via a three-clause 'Use when:' trigger list with concrete conditions.	5 / 5
Trigger Term Quality	Good keyword coverage including 'migrate', 'Model Serving', 'Apps', 'ResponsesAgent', 'predict()/predict_stream()', and '@invoke/@stream', though it lacks synonym variations a user might naturally phrase differently.	4 / 5
Distinctiveness Conflict Risk	A highly specific niche (MLflow ResponsesAgent migration from Model Serving to Apps) with distinct technical triggers means minimal overlap risk with unrelated skills.	5 / 5
	Total	18 / 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
skill_md_line_count	SKILL.md is long (968 lines); consider splitting into references/ and linking	Warning

	Total	15 / 16 Passed

Repository: databricks/app-templates
Path: agent-openai-agents-sdk/.claude/skills/migrate-from-model-serving/SKILL.md
Commit: fdc1b49

Reviewed: 1 day 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.