Component-aware code review that delegates to framework-specific skills. Reviews generated code against resolved guidelines, checks component patterns, and verifies LSP diagnostics are clean.
How this skill is triggered — by the user, by Claude, or both
Slash command
/frontend-coding-agent:code-reviewThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Runs after code generation and before final completion. Performs a structured review of all changed files, delegates framework-specific checks to the active framework module, and enforces LSP cleanliness.
Runs after code generation and before final completion. Performs a structured review of all changed files, delegates framework-specific checks to the active framework module, and enforces LSP cleanliness.
Execute the following steps in order:
Collect all files created or modified during the current session. Group by extension and directory to establish scope.
For each file, invoke the component-detection skill to classify the file as one of: page, component, hook, layout, utility, style. The component type drives which framework sub-skills are invoked and which guideline sections apply.
Invoke config-resolver to obtain the final merged guidelines for the current session. Guidelines are resolved from three layers (plugin defaults < team custom < per-invocation). The resolved output is a map of ## Section headings to their content.
For each changed file, verify it conforms to every applicable section in the resolved guidelines. A guideline violation is a blocker.
Invoke the active framework module's <fw>-components skill (e.g., react-components, angular-components, vue-components) for all files classified as page, component, layout, or hook. The framework skill checks framework-specific patterns (composition, lifecycle, props/inputs/outputs, state management integration).
Invoke the active framework module's <fw>-styling skill for all files that contain styling (CSS-in-JS, Tailwind classes, CSS Modules, SCSS). The styling skill verifies consistency with the resolved styling strategy and design token usage.
Run LSP.getDiagnostics on all changed files.
Run LSP.getReferences on all imports within changed files to verify that every import resolves to an existing module or export. Broken imports are blockers.
Invoke accessibility-review. This step always runs regardless of component type or configuration. Blockers reported by accessibility-review are blockers for code-review.
If i18n.enabled is true in the resolved config, invoke i18n-review. Blockers reported by i18n-review are blockers for code-review. If i18n.enabled is false or absent, skip this step.
Produce a structured review report grouped by severity, then by file.
## Code Review Report
### Blockers (must fix)
[BLOCKER] Guidelines — Violates "## Naming Conventions": component file uses kebab-case, must be PascalCase
File: src/components/user-profile.tsx, line 1
Guideline: ## Naming Conventions
Fix: Rename file to UserProfile.tsx
[BLOCKER] LSP — Type error: Argument of type 'string' is not assignable to parameter of type 'number'
File: src/components/UserProfile.tsx, line 28
Fix: Pass a numeric ID or update the function signature
[BLOCKER] Accessibility — Missing <main> landmark (see accessibility-review output)
### Warnings
[WARNING] Framework Patterns — useEffect dependency array missing 'userId' (react-components)
File: src/components/UserProfile.tsx, line 45
Fix: Add userId to the dependency array
[WARNING] Styling — Hardcoded color value; prefer design token (react-styling)
File: src/components/UserProfile.tsx, line 12
### Suggestions / Nits
[NIT] Guidelines — Consider extracting inline handler to named function for readability
File: src/components/UserProfile.tsx, line 67
The review is complete only when zero blockers remain. If blockers exist, report them and request fixes before marking the task done.
npx claudepluginhub gagandeepp/software-agent-teams --plugin frontend-coding-agentGuides completion of development work by verifying tests, detecting environment, and presenting structured options for merge, PR, or cleanup.
Enforces test-driven development: write failing test first, then minimal code to pass. Use when implementing features or bugfixes.
Guides creation and editing of skills using test-driven development with pressure scenarios and subagents to verify agent compliance.