The body is lean, assumes Claude knows Bazel, and uses a compact architecture tree plus a concept table instead of prose explanations, so every token earns its place; it is not at 2 because there is no unnecessary explanation to trim.

Do/don't bullets like 'Use fine-grained targets - Better caching' give some concrete guidance but the body contains no executable code or commands, delegating all templates to the reference; it is not at 1 because guidance is specific, and not at 3 because nothing in-body is copy-paste ready.

Content is organized as 'When to Use' and do/don't lists rather than a sequenced multi-step process, and there are no validation checkpoints or feedback loops; it is not at 1 because structure is present, but not at 3 because no real workflow sequence with validation exists.

A clear overview defers concrete templates to a well-signaled, one-level-deep, verified reference ('references/details.md'), keeping the body appropriately split and easy to navigate; it is not at 2 because the reference is clearly signaled and not deeply nested.

Names concrete actions such as 'Optimize Bazel builds', 'configuring Bazel', and 'implementing remote execution', matching the multiple-specific-actions anchor rather than the partial score-2 anchor.

It explicitly answers what ('Optimize Bazel builds for large-scale monorepos') and when ('Use when configuring Bazel, implementing remote execution, or optimizing build performance'), matching the explicit-trigger anchor.

Includes natural terms like 'configuring Bazel', 'remote execution', and 'build performance', but misses common variations a user might say such as 'build cache', 'remote cache', or 'Bazel build is slow', so it does not reach full coverage.

It carves a clear niche (Bazel build optimization for enterprise monorepos) with distinct triggers, making overlap with other skills unlikely; it is not below at 2 because it is not merely 'somewhat specific'.