Conciseness

The body is dense and opinionated ('A spec is in-flight scaffolding, not the durable record'; 'No process theater') and explains nothing Claude already knows, matching the score-4 anchor 'efficient; minor instances of over-explanation that could be trimmed'. The ~70-line field.* catalog example and the fully worked call-site example are longer than needed to convey the pattern, which is what keeps it from the lean score-5 anchor.

Actionability

Guidance is fully concrete and copy-paste ready: exact naming convention 'specs/YYYYMMDDThhmmss-feature-name.md', a decision-classification table with rules, markdown templates for every section with [PLACEHOLDER] markers, and a verbatim before/after call-site pattern with file:line references. This matches the score-5 anchor 'fully executable; copy-paste ready; specific examples cover the common cases'.

Workflow Clarity

The writing sequence is clear with most checkpoints present: apply the one-sentence-test before outlining ('If you can't name what this spec is about in one concrete sentence... the spec is not ready'), classify every material decision, record crystallized decisions as Proposed ADRs, and verify with Success Criteria checkboxes ('Tests pass / build succeeds'). It falls short of the score-5 anchor because the steps are distributed across prose sections rather than presented as one explicit ordered procedure with feedback loops for the drafting process itself.

Progressive Disclosure

Structure is good: an up-front References section with conditional on-demand loading ('Load these on demand based on the spec's decision surface'), and the one bundle reference, references/decision-hygiene.md, exists and is one level deep with clear trigger conditions. This matches the score-4 anchor 'good structure; most content is appropriately placed; references mostly clear; minor organization gaps' — the gaps being several sibling-skill cross-references (../writing-voice/SKILL.md, ../pull-request/references/body-patterns.md, ../rethink/references/clean-breaks.md, ../../../specs/README.md) that do not resolve within this bundle, and long inline worked examples that could live in references.

Specificity

The description names the domain and a single concrete action — 'Write technical specs that let agents implement autonomously' — which matches the anchor 'names domain and 1-2 concrete actions, but not comprehensive'. It does not list several specific actions (e.g., structure research findings, design decision tables, plan implementation phases), so it does not reach the score-4 anchor 'lists several specific actions'.

Completeness

It clearly answers 'what' ('Write technical specs that let agents implement autonomously') and 'when' with an explicit 'Use for' clause containing three concrete trigger phrases, matching the score-5 anchor 'clearly and explicitly answers both what AND when with concrete trigger phrases'.

Trigger Term Quality

Three quoted natural trigger phrases ('write a spec', 'plan this feature', 'create a planning doc') are exactly what a user would say, giving good keyword coverage. A few natural synonyms are missing (e.g., 'specification', 'design doc', 'RFC'), which keeps it below the comprehensive score-5 anchor.

Distinctiveness Conflict Risk

'Write technical specs' plus the specific quoted triggers carve out a mostly distinct niche from adjacent writing/planning skills. However, 'plan this feature' is a broad phrase that could overlap with general planning or architecture skills, so it does not fully meet the 'minimal conflict risk' of the score-5 anchor.