Conciseness

The body is dense with non-obvious platform facts (background scans disabled without CBUUID filters, MTU 23/20-byte default, restoration identifier keys) and avoids re-explaining BLE basics, so nearly every token earns its place. It falls just short of score 5 because the 'Core Communication Contract' points (.poweredOn, retaining CBPeripheral, UUID-filtered scans, canSendWriteWithoutResponse) are restated nearly verbatim in both 'Common Mistakes' and the 'Review Checklist' — deliberate reinforcement, but still trimmable.

Actionability

Guidance is concrete and executable in most respects — a copy-paste Info.plist XML snippet, exact API names (scanForPeripherals(withServices:options:), maximumWriteValueLength(for:)), and specific constants (CBCentralManagerOptionRestoreIdentifierKey, bluetooth-central). It sits at score 4 rather than 5 only because the body contains no Swift code for the most common case (setting up CBCentralManager and its delegate flow); that lives in the reference file.

Workflow Clarity

The numbered 'Core Communication Contract' establishes the critical ordering (wait for .poweredOn before scan/connect/advertise, retain before connect, check canSendWriteWithoutResponse before unacknowledged writes) and the 'Review Checklist' provides checkpoints, matching 'clear sequence with most checkpoints present'. No destructive/batch cap applies since BLE connection management is neither, but it is not score 5 because there is no explicit inline feedback loop for error recovery (e.g., on didDisconnect or failed writes) — those are delegated to the reference file.

Progressive Disclosure

The SKILL.md is a genuine overview: a Contents list plus a 'Route by Task' section whose seven links (e.g., references/ble-patterns.md#swiftui-ble-integration, #l2cap-channels, #write-flow-control) were verified to match real headings in the single 449-line bundle file, one level deep with no nested references. This matches the score-5 anchor 'clear overview with well-signaled one-level-deep references; content appropriately split; easy navigation'.

Specificity

The description enumerates multiple concrete capabilities — 'GATT discovery, state restoration, MTU sizing, and background modes' plus 'BLE scanning, peripheral connection, characteristic reads/writes/notifications, or peripheral advertisement' — covering both central and peripheral roles comprehensively. This matches the score-5 anchor 'lists multiple specific concrete actions; comprehensive coverage' rather than score 4, which expects minor coverage gaps.

Completeness

It explicitly answers both questions: what it does ('Build direct Bluetooth Low Energy central or peripheral workflows with Core Bluetooth, including GATT discovery, state restoration, MTU sizing, and background modes') and when to use it ('Use for BLE scanning, peripheral connection, characteristic reads/writes/notifications, or peripheral advertisement'). This matches the score-5 anchor with concrete trigger phrases; score 4 would require the 'when' clause to be less explicit, which it is not.

Trigger Term Quality

Natural trigger phrases like 'BLE scanning', 'peripheral connection', 'characteristic reads/writes/notifications', and 'peripheral advertisement' give good coverage of what users would actually say. A few natural variations are missing (e.g., 'Bluetooth' alone, 'BLE device', 'connect to a Bluetooth accessory'), so it falls just short of score 5's 'comprehensive coverage of natural terms including synonyms' while clearly exceeding score 3's 'missing common variations'.

Distinctiveness Conflict Risk

Naming 'Core Bluetooth', 'GATT', 'BLE', and both central/peripheral roles carves out a clear niche with distinct triggers, matching the score-5 anchor of minimal conflict risk. Score 4 ('minor overlap risk with closely related skills') fits less well since the description is anchored to one specific iOS framework and its unambiguous API vocabulary.