databricks-agent-bricks
Create and manage Databricks Agent Bricks: Knowledge Assistants (KA) for document Q&A, Genie Spaces for SQL exploration, and Supervisor Agents (MAS) for multi-agent orchestration. Use when building conversational AI applications on Databricks.
83
78%
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 ./databricks-skills/databricks-agent-bricks/SKILL.mdQuality
Discovery
85%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 description that clearly identifies three specific agent types within the Databricks ecosystem and provides an explicit 'Use when' trigger clause. The main weakness is that trigger terms could be broader to capture more natural user language variations (e.g., 'RAG', 'chatbot', 'document search'). The description is concise, uses third person voice, and carves out a clear niche.
Suggestions
Expand trigger terms to include natural user language variations such as 'RAG', 'chatbot', 'document search', 'data exploration', or 'agent workflow' to improve discoverability.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions: 'Create and manage Databricks Agent Bricks', with three distinct types enumerated — Knowledge Assistants for document Q&A, Genie Spaces for SQL exploration, and Supervisor Agents for multi-agent orchestration. | 3 / 3 |
Completeness | Clearly answers both 'what' (create and manage three types of Agent Bricks with specific purposes) and 'when' (Use when building conversational AI applications on Databricks), with an explicit 'Use when' clause. | 3 / 3 |
Trigger Term Quality | Includes some relevant keywords like 'Databricks', 'Agent Bricks', 'Knowledge Assistants', 'Genie Spaces', 'SQL exploration', 'multi-agent orchestration', and 'conversational AI'. However, it misses common user variations like 'chatbot', 'RAG', 'document search', 'agent framework', or abbreviations users might naturally use. | 2 / 3 |
Distinctiveness Conflict Risk | Highly distinctive due to the specific Databricks platform context and the named agent types (Knowledge Assistants, Genie Spaces, Supervisor Agents). Unlikely to conflict with generic AI or chatbot skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
72%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a solid skill document that provides comprehensive, actionable guidance for creating Databricks Agent Bricks with good progressive disclosure to related skills. Its main weaknesses are some verbosity in parameter documentation that could be tightened, and a workflow section that lacks explicit validation/verification steps for checking provisioning status and confirming brick functionality. The multi-modal supervisor agent example is excellent and highly actionable.
Suggestions
Add explicit validation steps to the workflow, such as polling endpoint status with manage_ka(action='get') and verifying ONLINE status before proceeding, with error recovery guidance for OFFLINE or stuck PROVISIONING states.
Trim the best practices section—items like 'use meaningful names' and 'provide descriptions' are generic advice Claude doesn't need; replace with domain-specific gotchas or constraints unique to Agent Bricks.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is reasonably efficient but includes some redundancy—the Genie Space section repeatedly defers to another skill while still listing partial details, and the best practices section contains generic advice Claude already knows (e.g., 'use meaningful names', 'provide descriptions'). The MCP tool parameter documentation is thorough but could be more compact. | 2 / 3 |
Actionability | The skill provides concrete tool calls with specific parameter names, values, and a fully executable multi-agent example with realistic arguments. The MCP tool documentation clearly specifies required vs optional parameters and exact action strings, making it copy-paste ready. | 3 / 3 |
Workflow Clarity | The typical workflow section outlines a clear 4-step sequence (generate data → create brick → wait for provisioning → add examples), but lacks explicit validation checkpoints—there's no guidance on checking endpoint status programmatically, handling provisioning failures, or verifying the brick works correctly before proceeding. | 2 / 3 |
Progressive Disclosure | Content is well-structured with a concise overview table, clear sections for each brick type, and well-signaled one-level-deep references to related skills and detailed pattern files (1-knowledge-assistants.md, 2-supervisor-agents.md, databricks-genie skill). Navigation between skills is explicit with relative links. | 3 / 3 |
Total | 10 / 12 Passed |
Validation
100%Checks the skill against the spec for correct structure and formatting. All validation checks must pass before discovery and implementation can be scored.
Validation — 11 / 11 Passed
Validation for skill structure
No warnings or errors.
- Repository
- databricks-solutions/ai-dev-kit
- Commit
02aac8c
- 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.