plugins/ai-pilot/skills/code-quality/SKILL.md
Code quality validation with linters, SOLID principles, DRY detection, error detection, and architecture compliance across all languages. Use when: validating code quality after modifications — SOLID compliance, DRY duplication, linter errors, or architecture violations. Do NOT use for: verifying the original problem is functionally resolved (run verification FIRST — code-quality validates quality AFTER functional verification passes).
npx skillsauth add fusengine/agents code-qualityInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Canonical workflow definition: agents/sniper.md (this skill and the sniper agent share the same 7-phase workflow — update both together).
PHASE 1: Exploration (explore-codebase) → BLOCKER
PHASE 2: Documentation (research-expert) → BLOCKER
PHASE 3: Impact Analysis (Grep usages) → BLOCKER
PHASE 3.5: DRY Detection (jscpd duplication) → NON-BLOCKING
PHASE 4: Error Detection (linters)
PHASE 5: Precision Correction (with docs + impact + DRY)
PHASE 6: Verification (re-run linters, tests, duplication)
CRITICAL: Phases 1-3 are BLOCKERS. Never skip them. DRY: Phase 3.5 is non-blocking but findings inform Phase 5 corrections.
Launch explore-codebase agent FIRST:
> Agent(subagent_type="fuse-ai-pilot:explore-codebase", prompt="...")
Gather:
Launch research-expert agent:
> Agent(subagent_type="fuse-ai-pilot:research-expert", prompt="Verify [library/framework] documentation for [error type]. Find [language] best practices for [specific issue].")
Request for each error:
For EACH element to modify: Grep usages → assess risk → document impact.
| Risk | Criteria | Action | |------|----------|--------| | 🟢 LOW | Internal, 0-1 usages | Proceed | | 🟡 MEDIUM | 2-5 usages, compatible | Proceed with care | | 🔴 HIGH | 5+ usages OR breaking | Flag to user FIRST |
Tool: jscpd — 150+ languages — npx jscpd ./src --threshold 5 --reporters console,json
| Level | Threshold | Action | |-------|-----------|--------| | 🟢 Excellent | < 3% | No action needed | | 🟡 Good | 3-5% | Document, fix if time | | 🟠 Acceptable | 5-10% | Extract shared logic | | 🔴 Critical | > 10% | Mandatory refactoring |
See references/duplication-thresholds.md for per-language thresholds, config, and extraction patterns. See references/linter-commands.md for language-specific jscpd commands.
See references/linter-commands.md for language-specific commands.
| Priority | Type | Examples | Action | |----------|------|----------|--------| | Critical | Security | SQL injection, XSS, CSRF, auth bypass | Fix IMMEDIATELY | | High | Logic | SOLID violations, memory leaks, race conditions | Fix same session | | High | DRY | Code duplication > 10%, copy-paste logic blocks | Mandatory refactoring | | Medium | DRY | Code duplication 5-10%, repeated patterns | Extract shared logic | | Medium | Performance | N+1 queries, deprecated APIs, inefficient algorithms | Fix if time | | Low | Style | Formatting, naming, missing docs | Fix if time |
See references/solid-validation.md for S-O-L-I-D detection patterns and fix examples.
See references/file-size-rules.md for LoC limits, calculation, and split strategies.
See references/architecture-patterns.md for project structures and patterns.
See references/validation-report.md for the complete sniper report template.
See references/examples.md for detailed walkthrough.
testing
Copy self-audit and ban-lists — filler verbs/hype adjectives, slop placeholder names, fake-precise numbers, Title Case headlines, humor in error copy ('Oops!'), em-dash crutch, one copy register per page.
development
Logged-in web apps — dashboards, auth flows, settings, onboarding, data tables, command palettes, modals, toasts. Register `product`: density and glance-speed over marketing polish, no hero/CTA-tricks, every data surface covers empty/loading/error explicitly, tables and dataviz follow preattentive-processing rules.
development
Marketing sites, landing pages, campaign pages — register `brand` (design IS the product). Structure comes from the register's POV + a macrostructure pick, never from copying an inspiration site's section flow. Hero discipline, deviated section order, asymmetric grids, and a silhouette lookalike-test gate before ship.
development
Token-strategy core — OKLCH color rules, neutral tinting, accent-commitment levels, type scale, 8pt spacing grid, touch targets, and the canonical output format of design-system.md (the file the harness gates on). This is routing step 1 of design-method/SKILL.md — read it before design-web/design-webapp/design-ios/design-android, before picking or auditing a single color/type/spacing value.