databricks-iceberg
Apache Iceberg tables on Databricks — Managed Iceberg tables, External Iceberg Reads (fka Uniform), Compatibility Mode, Iceberg REST Catalog (IRC), Iceberg v3, Snowflake interop, PyIceberg, OSS Spark, external engine access and credential vending. Use when creating Iceberg tables, enabling External Iceberg Reads (uniform) on Delta tables (including Streaming Tables and Materialized Views via compatibility mode), configuring external engines to read Databricks tables via Unity Catalog IRC, integrating with Snowflake catalog to read Foreign Iceberg tables
76
93%
Does it follow best practices?
Impact
—
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 an excellent skill description that thoroughly covers a well-defined technical niche. It lists specific capabilities, includes abundant natural trigger terms that practitioners would use, provides an explicit 'Use when' clause with multiple concrete scenarios, and occupies a clearly distinct domain. The description is comprehensive without being padded with fluff.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists multiple specific concrete actions and technologies: creating Iceberg tables, enabling External Iceberg Reads on Delta tables, configuring external engines via Unity Catalog IRC, integrating with Snowflake catalog, credential vending, PyIceberg, OSS Spark access. | 3 / 3 |
Completeness | Clearly answers both 'what' (Apache Iceberg tables on Databricks with specific capabilities listed) and 'when' (explicit 'Use when...' clause with four distinct trigger scenarios covering table creation, External Iceberg Reads, external engine configuration, and Snowflake integration). | 3 / 3 |
Trigger Term Quality | Excellent coverage of natural terms a user would use: 'Iceberg tables', 'Databricks', 'Uniform', 'Delta tables', 'Streaming Tables', 'Materialized Views', 'compatibility mode', 'Unity Catalog', 'IRC', 'Snowflake', 'PyIceberg', 'OSS Spark', 'credential vending', 'Iceberg v3'. These are the exact terms practitioners would use. | 3 / 3 |
Distinctiveness Conflict Risk | Highly distinctive niche combining Apache Iceberg + Databricks + specific interop scenarios. The domain-specific terminology (IRC, Unity Catalog, External Iceberg Reads, credential vending) makes it very unlikely to conflict with other skills. | 3 / 3 |
Total | 12 / 12 Passed |
Implementation
87%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This is a high-quality skill that efficiently covers a complex topic with multiple interrelated features. The critical rules section is exceptionally valuable, capturing non-obvious gotchas (metadata path corruption, library conflicts, DV/row-tracking requirements) that would be hard to discover otherwise. The main weakness is the lack of explicit validation steps in workflows — particularly for operations like enabling UniForm or creating Iceberg tables where verification would prevent silent failures.
Suggestions
Add a brief validation step after the UniForm ALTER TABLE example (e.g., `DESCRIBE EXTENDED table_name` to confirm Iceberg metadata generation status), especially since async delay is listed as a common issue.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The content is dense with domain-specific knowledge that Claude wouldn't inherently know — Iceberg-specific TBLPROPERTIES, version-specific behaviors, UniForm nuances, and critical gotchas. There's no padding or explanation of basic concepts. Every section earns its place. | 3 / 3 |
Actionability | The Quick Start provides fully executable SQL examples covering multiple scenarios (no clustering, PARTITIONED BY, CLUSTER BY on v2 vs v3, UniForm enablement). The Common Issues table gives specific solutions with exact property names and commands. The capability matrix is immediately useful for decision-making. | 3 / 3 |
Workflow Clarity | The Critical Rules section clearly states constraints and the When to Use section routes users well, but there are no explicit validation checkpoints or feedback loops. For example, after enabling UniForm there's no 'verify it worked' step, and the ALTER TABLE workflow doesn't mention checking async metadata generation status despite noting it as a common issue. | 2 / 3 |
Progressive Disclosure | Excellent structure: concise overview with critical rules and quick start at the top, then a well-organized reference table with clear file descriptions and keywords, followed by a use-case-driven navigation section. All references are one level deep and clearly signaled with descriptive summaries. | 3 / 3 |
Total | 11 / 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
93cb4e3
- 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.