Names the domain (dbt documentation) and some actions ('retrieves and searches', 'LLM-friendly markdown format'), but doesn't list multiple specific concrete actions like extracting code examples, finding configuration syntax, or comparing versions.

Clearly answers both what ('Retrieves and searches dbt documentation pages in LLM-friendly markdown format') and when ('Use when fetching dbt documentation, looking up dbt features, or answering questions about dbt Cloud, dbt Core, or the dbt Semantic Layer').

Includes strong natural keywords users would say: 'dbt documentation', 'dbt features', 'dbt Cloud', 'dbt Core', 'dbt Semantic Layer'. These cover the main product variants users would mention.

Very clear niche focused specifically on dbt documentation with distinct product names (dbt Cloud, dbt Core, dbt Semantic Layer). Unlikely to conflict with general documentation or other data tool skills.

The content is lean and efficient, providing only essential information without explaining concepts Claude already knows. Tables are used effectively to compress information, and there's no unnecessary padding or explanation of what dbt or documentation is.

Provides concrete, executable examples including exact URL patterns, bash commands with real examples, and copy-paste ready WebFetch commands. The script usage examples show multiple use cases with actual keywords.

Clear two-step workflow for finding pages (search index first, then full docs only if needed). The sequence is explicit with 'Step 1' and 'Step 2' labels, and the 'Common Mistakes' table serves as a validation checklist to catch errors.

Well-organized with clear sections progressing from overview to quick reference to detailed usage. Content is appropriately contained in a single file given its scope, with external script reference clearly signaled. No unnecessary nesting or monolithic walls of text.