Conciseness

The body is dense and assumes competence — a routes table, terse failure/recovery bullets, and a copy-ready CLI block with no explanation of concepts Claude already knows. It falls short of anchor 5 only because of small trimmable meta-content (e.g. "persistent caches belong in generated state, not skill definitions") and a few opaque phrases ("material fallback limits").

Actionability

Concrete, executable guidance dominates: `oma search code "<query>" [--host gitlab] [--language <lang>]`, `oma search trust <domain>`, `oma search fetch <url>`, and named tool calls (Context7 `resolve-library-id` → `query-docs`). It is not anchor 5 because the docs and web routes stop at tool names — the actual Context7 invocation and runtime web-search call patterns are deferred or unstated.

Workflow Clarity

The four-step canonical workflow is clearly sequenced (parse → dispatch → collect/dedupe/trust-score → rank and return), backed by an explicit failure-and-recovery section covering fallbacks for each failure mode. It misses anchor 5 because validation checkpoints are described only at route-failure level and detail is deferred to `resources/checklist.md` and `resources/error-playbook.md` rather than stated inline.

Progressive Disclosure

The body is a genuine overview: a dedicated References section lists each resource file with its purpose and load condition (e.g. "Intent classifier: `resources/intent-rules.md` (no explicit route or ambiguous query)"), references are one level deep, and workflow steps point to detail files only when needed. No bundle files were provided to verify, but the structure itself matches the clear-overview, well-signaled-references anchor.

Specificity

The description names its domain ("external documentation, web sources, or remote code") and one concrete action ("Find ... with citations") plus a boundary sentence, which matches the anchor for a domain with 1-2 concrete actions but not comprehensive coverage. It does not list several specific actions such as resolving library docs, fetching URLs, or ranking sources, so it falls below anchor 4.

Completeness

The "what" is clear (find external docs/web/remote code with citations), but there is no "Use when..." clause or equivalent explicit trigger guidance; the second sentence is a negative boundary ("Local code navigation uses the configured code tools") that only weakly implies when to use it. Per the judging guidelines a missing explicit trigger caps completeness at 3, and it is not anchor 2 because the "what" is concrete.

Trigger Term Quality

Phrases like "documentation", "web sources", "remote code", "citations", and "Local code navigation" are relevant keywords, but common natural variations users would say — "search", "look up", "API reference", "research" — are absent. This lands on the anchor for some relevant keywords missing common variations rather than the good-coverage anchor 4.

Distinctiveness Conflict Risk

The description carves out a clear niche — external/web/remote search — and explicitly disambiguates the nearest conflict ("Local code navigation uses the configured code tools"), giving it distinct triggers with minimal conflict risk. Anchor 4 would require some overlap concern, which the explicit boundary removes.