Conciseness

The body is dense and assumes Claude's competence (no explanations of uv, pytest, or general concepts; every gate rule is referenced by id rather than restated). It falls short of the 5 anchor because the ~20-line "Maintaining the gate" section — config notes, glob syntax, pytest invocation — serves a maintainer rather than a code-change audience and could be trimmed or moved to a reference; it is noticeably above the 3 anchor since the rest is lean and padded nowhere.

Actionability

Guidance is fully executable: three copy-paste-ready invocations of the gate (``uv run python .agents/skills/kiln-conventions/scripts/conventions_gate.py --worktree`` / ``--range`` / ``--files``), the exact output format (``SEV<TAB>RULE<TAB>path:line<TAB>snippet``), exit-code semantics, per-severity handling instructions, and a concrete example WARN justification. It matches the 5 anchor (copy-paste ready commands covering the common cases).

Workflow Clarity

A clear 5-step sequence with explicit validation (step 4 runs the gate on the change; exit 1 means a FAIL), feedback loops for error recovery (fix and re-run, allowlist with a reason line, or report an unfixable FAIL with rule id and path:line), and a checklist for what the gate cannot see (step 5's rule-by-rule self-check), plus a separate reviewer workflow. This matches the 5 anchor's "explicit validation steps; feedback loops; checklists".

Progressive Disclosure

The body is a workflow overview with well-signaled, one-level-deep references: a path-to-reference table tells the agent to read only ``references/rules.md`` plus the per-area file matching the edited paths (all four reference files and all cited scripts exist, and every rule id cited in the body is defined in rules.md). Content is appropriately split between SKILL.md, references, and scripts, and navigation is easy; the inline maintainer section is clearly labeled and too minor to drop this to the 4 anchor.

Specificity

The description enumerates concrete capabilities — "Kiln's code conventions and per-area gotchas, plus a diff gate that enforces the mechanical ones" — and lists coverage topics ("comments, startup and entry points, globals and module state, config reads, module layering, library-vs-app rules and async I/O"), leaving only minor gaps. It falls short of the 5 anchor because it names content areas rather than multiple specific executable actions like "extract/fill/merge".

Completeness

It explicitly answers both questions: the "what" is "code conventions and per-area gotchas, plus a diff gate that enforces the mechanical ones" with a topic list, and the "when" is a concrete trigger clause ("Invoke before writing or changing any Python, TypeScript or Svelte code… and when reviewing such a change"). This matches the 5 anchor (clear what AND when with concrete trigger phrases); anchor 4 would require the "when" to be less explicit.

Trigger Term Quality

Natural trigger phrases are present: "Invoke before writing or changing any Python, TypeScript or Svelte code in the Kiln repo… and when reviewing such a change", plus concrete paths (libs/core, libs/server, app/desktop, app/web_ui). A few natural variations users might say ("editing", "refactoring" Kiln code) are missing, so it fits the "good keyword coverage; a few natural terms missing" anchor rather than the comprehensive 5 anchor.

Distinctiveness Conflict Risk

The description is scoped to a named repo ("in the Kiln repo") with specific sub-paths and languages, giving it a clear niche with distinct triggers and minimal conflict risk with other skills. It is well above the 4 anchor ("minor overlap risk with closely related skills") because the repo and path scoping make false triggers very unlikely.