About inspect-role-naming
inspect-role-naming is a skill for auditing architectural roles and naming in a codebase against repository naming standards. It distinguishes localized renames from structural refactors and produces a findings dossier.
Key features
- 01Auditing names against a governing rubric
- 02Separating rename and refactor findings
- 03Ignoring cosmetic or low-risk issues
- 04Producing a structured audit dossier
Use cases
- 01Reviewing architecture-related naming drift
- 02Identifying symbols with misleading responsibilities
- 03Routing local fixes versus structural changes
Install
Copy the command and run it in a terminal.
$
npx skills add https:/ / github.com/ venhdev/ skills --skill inspect-role-namingAfter installing, type/inspect-role-namingin Claude Code to confirm it loaded.
Bundled filesreferences/role-naming-rubric.md
Details
From the project READMEinspect-role-naming — Architectural Role & Nomenclature Auditor
Domain Rubric
1. Scope & Tolerability Guardrails (What to Ignore)
- Confine to Load-Bearing Risks: Focus exclusively on dangerous architectural misrepresentations, runtime collisions, or boundary schema leaks.
- Tolerate Benign Code: Forgive minor local helpers (< 3 methods), internal variable brevity, benign metrics/logging in accessors, and cosmetic style differences. Forbid forcing refactoring on working, cohesive code.
- Governing Reference: Evaluate naming and role alignment strictly against
references/role-naming-rubric.md(or the host repository's authoritative naming specifications, which always take precedence).
2. Decision Bifurcation: [RENAME] vs. [REFACTOR]
[RENAME](Pure Nomenclature Drift): The underlying responsibility is singular and sound; only the identifier/suffix drifted fromreferences/role-naming-rubric.md. Fix is localized and low-risk.[REFACTOR](Role Dishonesty / Responsibility Leak): The symbol name conceals multiple disparate responsibilities or makes false promises (e.g., repository calling external APIs). Structural extraction must occur before renaming.
Canonical Output Contract
# Role & Naming Audit Dossier: <Target Scope>
## 1. Executive Summary
- **Target Inspected**: `<directory_or_file_path>`
- **Health Rating**: <PRISTINE | LOCALIZED_DRIFT | SYSTEMIC_DEBT>
- **Primary Bottleneck**: <Single-sentence summary of greatest role or boundary hazard>
## 2. Load-Bearing Defect Catalog
### [<CRITICAL | MAJOR>] [<RENAME | REFACTOR>] <Dimension>: `<Target Symbol>`
- **Offending Code**: [<file>#L<N>](file:///path/to/file#L<N>)
- **Violated Rubric**: [<spec_or_rubric>](references/role-naming-rubric.md#anchor)
- **Smell**: <Observable contract mismatch or dangerous collision>
- **Action**: <[RENAME] New Identifier OR [REFACTOR] Minimal Separation Path>
*(Emit 'No load-bearing role or naming defects observed' if clean)*
Routing
- Localized symbol renames or minor boolean cleanup within a single file → (via /simplify-audit).
- Multi-file symbol migrations or structural role extractions → (via /changeset).
Execution Protocol
-
Intake & Reference Grounding:
- Ingest target path (
directoryorfiles). - Ground in the host repository's naming standards if present (project authority takes priority); otherwise read and apply
references/role-naming-rubric.md.
- Ingest target path (
-
In-Turn Pragmatic Audit:
- Audit target code directly in-turn against rubric dimensions.
- Enforce Section 1 Tolerability Guardrails: strictly ignore benign idioms, local brevity, and working helpers.
- For every finding, classify the remediation explicitly as either
[RENAME]or[REFACTOR].
-
Synthesis & Turn-Halt Gate:
- Consolidate findings into the Canonical Output Contract.
- Assign definitive Health Rating (
PRISTINE,LOCALIZED_DRIFT, orSYSTEMIC_DEBT). - Present completed dossier into conversation stream and halt turn immediately.
Comments
Markdown supported