Conciseness

The same examples appear repeatedly — `{"confidence": {"$gt": 0.7}}` is shown in Comparisons, the Quick Reference table, the bm_search section, and the Custom Frontmatter example, and the Guidelines section restates rules already given ("Multiple filters are AND", "Omit `query` for filter-only searches", "Dot notation for nesting"). This is more than the minor over-explanation of anchor 4 but well short of the concept-explaining padding of anchor 2, so it sits at anchor 3: mostly efficient but could be tightened.

Actionability

Fully executable copy-paste Python throughout (`bm_search(filters={"status": "in-progress"}, project="<project>")`), a complete JSON filter syntax with every operator, a quick-reference table, and explicit edge-case rules (non-empty lists, `$between` bounds, null semantics, key regex). The `<project>` placeholder is explained up front, so examples run as-is once substituted.

Workflow Clarity

A simple single-purpose skill where the action is unambiguous ("Use `bm_search` for ranked results or `bm_find` for file discovery... Omit `query`... for filter-only searches"), which qualifies for the simple-skill exception. The project-scoping mini-workflow is sequenced with edge handling — discover via `bm_project(command="list")`, keep scope when "following returned paths, paginating, or recovering from a refusal" — and nothing here is destructive or batch, so no validation cap applies.

Progressive Disclosure

No bundle files exist, and the body is well organized with clear sections (When to Use, The Tool, Filter Syntax, Quick Reference, Guidelines) plus a one-level pointer to the memory-notes skill for observation qualifiers. It misses anchor 5 because at ~220 lines (over the 50-line simple-skill threshold) the full operator reference and the worked example section are inlined where a reference file could carry the bulk; structure is good but the split is not optimal.

Specificity

"Search Basic Memory notes by custom frontmatter fields with equality, range, array, and nested filters" names the domain plus several concrete capabilities (equality, range, array, nested filters). It falls short of anchor 5 because it omits capabilities the body covers (tag shorthand, bm_find field discovery, combining text + metadata), but exceeds anchor 3 since more than 1-2 concrete actions are enumerated.

Completeness

Both halves are explicit: the what is "Search Basic Memory notes by custom frontmatter fields with equality, range, array, and nested filters" and the when is "Use when finding notes by status, priority, confidence, or any YAML field, not free text" with concrete trigger phrases. This matches the anchor-5 example structure exactly; anchor 4 would require the when to be less explicit or specific, which it is not.

Trigger Term Quality

"Use when finding notes by status, priority, confidence, or any YAML field, not free text" supplies good natural trigger terms users would say. Not anchor 5: common synonyms and variations such as "tags", "metadata", "custom fields", or "by field value" are only partially covered ("custom frontmatter fields").

Distinctiveness Conflict Risk

"Basic Memory notes" plus the explicit carve-out "not free text" establishes a clear niche and actively disambiguates from free-text search skills. Minimal conflict risk; the trigger conditions (status, priority, confidence, YAML field) are specific to structured metadata lookup.