refactor(cc-lint): remove stale Portability dimension, fold into Correctness

Portability was really just field conformance (use only documented
frontmatter fields) — not cross-platform portability. Fold it into
Correctness where it belongs, add trigger synonyms, align approach
steps with evaluation-process.md, and add conflict resolution guidance.

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>

Author: philoserf

skills/cc-check/references/common-failures.md

@@ -24,16 +24,15 @@ Reference this file when debugging test failures. Each pattern includes symptoms
 
 **Fix**: Add trigger phrase variations to description. Include both formal terms and casual language users might use.
 
-### Portability Issue
+### Field Conformance Issue
 
-**Symptom**: Skill uses non-standard frontmatter fields or agent-specific coupling
+**Symptom**: Skill uses non-standard frontmatter fields
 
 **Common causes**:
 
 - Non-standard fields in frontmatter (beyond documented fields)
-- Instructions tightly coupled to a specific implementation
 
-**Diagnosis**: Check frontmatter for documented fields only.
+**Diagnosis**: Check frontmatter against documented fields: `name`, `description`, `argument-hint`, `disable-model-invocation`, `user-invocable`, `allowed-tools`, `model`, `effort`, `context`, `agent`, `hooks`.
 
 **Fix**: Remove non-standard frontmatter fields.
 

skills/cc-lint/SKILL.md

@@ -1,7 +1,7 @@
 ---
 disable-model-invocation: true
 allowed-tools: Read, Grep, Glob, Bash
-description: Performs quick structural validation of Claude Code customizations. Use when linting, validating structure, checking frontmatter, or reviewing any skill, agent, or hook for correctness. Checks YAML frontmatter, required fields, naming conventions, file organization, and settings.json health.
+description: Performs quick structural validation of Claude Code customizations. Use when linting, validating structure, checking frontmatter, auditing customizations, reviewing skill structure, or reviewing any skill, agent, or hook for correctness. Checks YAML frontmatter, required fields, naming conventions, file organization, and settings.json health.
 ---
 
 ## Reference Files
@@ -41,16 +41,13 @@ If a specific file or directory is passed as an argument, lint that target direc
 
 ## Approach
 
-When evaluating a Claude Code customization, this skill follows a systematic process:
+When evaluating a Claude Code customization, this skill follows the process in [evaluation-process.md](references/evaluation-process.md):
 
-1. Read and parse target file(s) to extract structure and content
-2. Validate YAML frontmatter for documented fields and correct syntax
-3. Apply type-specific validation criteria (agent/skill/hook)
-4. Assess context economy and progressive disclosure usage
-5. Verify only documented frontmatter fields are used
-6. Check integration with settings.json and other customizations
-7. Generate structured report with specific findings and recommendations
-8. Prioritize issues by severity (correctness > clarity > effectiveness)
+1. Determine scope and identify extension type
+2. Apply type-specific validation (frontmatter, structure, fields)
+3. Check integration with settings.json
+4. Assess context economy and progressive disclosure
+5. Generate structured report with prioritized findings
 
 Detailed criteria, process steps, and examples are available in the reference files above.
 

skills/cc-lint/assets/report-format.md

@@ -20,7 +20,6 @@ This document defines the standardized structure for evaluation reports.
 - **Correctness**: PASS | NEEDS WORK | FAIL
 - **Clarity**: PASS | NEEDS WORK | FAIL
 - **Effectiveness**: PASS | NEEDS WORK | FAIL
-- **Portability**: PASS | NEEDS WORK | N/A
 
 ## Correctness Findings
 
@@ -52,16 +51,6 @@ This document defines the standardized structure for evaluation reports.
 - {issue 2}
 - ...
 
-## Portability Findings
-
-{Field conformance and structural quality issues}
-
-**Issues Found**: {count}
-
-- {issue 1}
-- {issue 2}
-- ...
-
 ## Context Usage
 
 - **File Size**: {bytes or "N/A for multiple files"}

skills/cc-lint/references/evaluation-criteria.md

@@ -5,9 +5,11 @@ This document defines the correctness, clarity, and effectiveness standards for
 ## Contents
 
 - [Correctness Criteria](#correctness-criteria) — Agents, Skills, Hooks
-- [Clarity Criteria](#clarity-criteria) — Description quality, structure, portability
+- [Clarity Criteria](#clarity-criteria) — Description quality, structure
 - [Effectiveness Criteria](#effectiveness-criteria) — Context economy, triggering, integration
 
+When dimensions conflict (e.g., brevity aids context economy but hurts trigger coverage), prioritize correctness first, then effectiveness, then clarity.
+
 ## Correctness Criteria
 
 ### Agents
@@ -37,7 +39,7 @@ Per the [Claude Code skills docs](https://docs.anthropic.com/en/docs/claude-code
 
 - All fields are optional; only `description` is recommended
 - Documented fields: `name`, `description`, `argument-hint`, `disable-model-invocation`, `user-invocable`, `allowed-tools`, `model`, `effort`, `context`, `agent`, `hooks`
-- Any other field is non-standard — flag as a warning
+- Any other field is non-standard — flag as a warning (field conformance)
 
 **Structure validation**:
 
@@ -70,14 +72,6 @@ Per the [Claude Code skills docs](https://docs.anthropic.com/en/docs/claude-code
 - References used when content exceeds ~500 lines
 - References clearly linked from SKILL.md, one level deep
 
-### Portability (Skills)
-
-Per the [Claude Code skills docs](https://docs.anthropic.com/en/docs/claude-code/skills):
-
-- Only documented frontmatter fields used (`name`, `description`, `argument-hint`, `disable-model-invocation`, `user-invocable`, `allowed-tools`, `model`, `effort`, `context`, `agent`, `hooks`)
-- No unnecessary assumptions baked into structure
-- Tool names documented as implementation details, not hard requirements
-
 ## Effectiveness Criteria
 
 ### Context Economy

skills/cc-lint/references/evaluation-process.md

@@ -69,14 +69,6 @@ Calculate approximate size and efficiency:
 3. Check progressive disclosure usage
 4. Estimate token count impact
 
-## Step 5: Assess Portability
-
-For skills, confirm only documented frontmatter fields are used (verified in Step 2) and assess broader quality:
-
-1. Check for non-standard frontmatter fields
-2. Assess whether content is clear and well-structured
-3. Note any non-documented fields as warnings
-
-## Step 6: Generate Structured Report
+## Step 5: Generate Structured Report
 
 Create comprehensive evaluation following the report format (see [report-format.md](../assets/report-format.md#report-template)).