Conciseness

Lean and dense throughout — the overview, cheat sheet, and anti-patterns each carry actionable signal with no explanations of concepts Claude already knows; every token earns its place.

Actionability

The path cheat sheet gives copy-paste-ready commands (cat ~/.deeplake/memory/graph/query/<pattern>, impact/<symbol>, path/<from>/<to>) with concrete expected output comments covering the common query cases.

Workflow Clarity

The numbered Workflow gives a clear decision-driven sequence (broad→find→show→path/impact→Read source), and the Anti-patterns section supplies validation checkpoints (confirm "Incoming (0)" in source, check build age before trusting stale graph) for this read-only skill.

Progressive Disclosure

No bundle files exist and none are needed; the body is self-contained, well-split into clearly-headed sections (When to use / NOT to use / cheat sheet / Workflow / Anti-patterns) with easy header-based navigation and no nested references.

Specificity

Lists multiple concrete capabilities — querying "functions, classes, calls, imports" plus transitive "impact of changing this" — giving comprehensive coverage of the graph's query surface rather than vague abstraction.

Completeness

Explicitly answers both what ("Query the local code graph ... AST-derived map of the repo") and when ("Use when the user asks structural questions about the codebase") with concrete trigger phrases.

Trigger Term Quality

Includes the exact natural phrases users say — "what calls X?", "what does Y import?", "where is Z defined?", "what's the architecture / which subsystems exist?", "what's the impact of changing this?" — covering synonyms and variations comprehensively.

Distinctiveness Conflict Risk

Occupies a clear niche — structural/relational code questions answered via a specific Deeplake graph mount — with distinct triggers and minimal overlap risk against generic code-search or read skills.