warehouse-init
Initialize warehouse schema discovery. Generates .astro/warehouse.md with all table metadata for instant lookups. Run once per project, refresh when schema changes. Use when user says "/astronomer-data:warehouse-init" or asks to set up data discovery.
68
60%
Does it follow best practices?
Impact
Pending
No eval scenarios have been run
Risky
Do not use without reviewing
Optimize this skill with Tessl
npx tessl skill review --optimize ./skills/warehouse-init/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 well-crafted description that clearly communicates what the skill does (generates warehouse metadata files), when to use it (explicit slash command and natural language trigger), and its operational context (run once, refresh on schema changes). The main weakness is limited natural language trigger terms—users asking about database schemas, table exploration, or catalog setup might not trigger this skill.
Suggestions
Expand trigger terms to include natural variations like 'explore schema', 'list tables', 'database catalog', or 'discover tables' to improve matching for users who don't know the slash command.
| Dimension | Reasoning | Score |
|---|---|---|
Specificity | Lists specific concrete actions: 'Initialize warehouse schema discovery', 'Generates .astro/warehouse.md with all table metadata for instant lookups'. Describes what is produced and its purpose clearly. | 3 / 3 |
Completeness | Clearly answers both what ('Generates .astro/warehouse.md with all table metadata for instant lookups') and when ('Use when user says "/astronomer-data:warehouse-init" or asks to set up data discovery', plus 'Run once per project, refresh when schema changes'). | 3 / 3 |
Trigger Term Quality | Includes some relevant terms like 'warehouse', 'schema', 'data discovery', and the slash command trigger '/astronomer-data:warehouse-init'. However, it misses natural variations users might say like 'database tables', 'catalog', 'explore schema', or 'list tables'. | 2 / 3 |
Distinctiveness Conflict Risk | Highly distinctive with a specific slash command trigger, a unique output artifact (.astro/warehouse.md), and a clear niche around warehouse schema discovery. Unlikely to conflict with other skills. | 3 / 3 |
Total | 11 / 12 Passed |
Implementation
35%Reviews the quality of instructions and guidance provided to agents. Good implementation is clear, handles edge cases, and produces reliable results.
This skill is comprehensive but far too verbose for its purpose, violating token efficiency principles. It contains useful concrete commands and a clear multi-step workflow, but mixes actionable instructions with explanatory content that Claude doesn't need (cache staleness signs, what dbt models are). The lack of validation checkpoints in a multi-step process involving warehouse queries and file generation is a notable gap.
Suggestions
Cut content by 50%+: Remove the example session (duplicates the process), trim the output format template to just key structural elements, and eliminate explanatory sections like 'Signs of Stale Cache' and 'Codebase Patterns Recognized' that Claude can infer.
Add validation checkpoints: After Step 3 (warehouse discovery), verify that queries returned results before proceeding. After Step 6 (generate warehouse.md), validate the file was written and contains expected sections.
Fix inconsistent CLI paths: Standardize all commands to use the same path format (either 'uv run scripts/cli.py' or 'uv run cli.py') and clarify the working directory once at the top rather than repeating path notes.
Split reference content into separate files: Move the full output format template to a TEMPLATE.md file and the cache management/refresh details to a MAINTENANCE.md file, keeping SKILL.md as a lean process guide.
| Dimension | Reasoning | Score |
|---|---|---|
Conciseness | The skill is extremely verbose at ~250+ lines. It over-explains concepts Claude can infer (e.g., what schema discovery does, signs of stale cache, what dbt models are), includes lengthy output format templates, and repeats information (e.g., script paths mentioned multiple times, the Quick Reference table appears in multiple sections). The example session at the end largely duplicates the process steps. | 1 / 3 |
Actionability | Provides concrete CLI commands and SQL queries, but many steps use pseudocode-like Task() calls that aren't real executable syntax. The subagent patterns are illustrative rather than copy-paste ready. Some commands reference relative paths inconsistently (e.g., 'uv run cli.py' vs 'uv run scripts/cli.py'). | 2 / 3 |
Workflow Clarity | The 8-step process is clearly sequenced and logically ordered, but lacks explicit validation checkpoints. There's no verification that warehouse.md was generated correctly, no error handling for failed SQL queries or connection issues, and no feedback loop if discovery fails for a database. The refresh behavior section describes preservation logic but doesn't validate the merge. | 2 / 3 |
Progressive Disclosure | The content is mostly monolithic — everything is in one long file with no references to separate documentation. Sections like 'Codebase Patterns Recognized', 'Cache Staleness & Schema Drift', and the full output format template could be split into referenced files. The structure has clear headings but the sheer volume undermines navigability. | 2 / 3 |
Total | 7 / 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
- astronomer/agents
- Commit
0642adb
- 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.