Install
$ agentstack add mcp-piyush8296-claude-workspace Open-source listing β not yet scanned by AgentStack. Follow the source repository for install instructions.
Security review
β Flagged1 finding(s); flagged for manual review. Β· v0.1.0 How review works β
- β’ Prompt-injection patterns
- β’ Secret / credential exfiltration
- β’ Dangerous shell & filesystem operations
- β’ Untrusted network calls
- β’ Known-malicious package signatures
- high Pipes remote content directly into a shell (remote code execution).
What it can access
- β Network access Used
- β Filesystem access No
- β Shell / process execution No
- β Environment & secrets Used
- β Dynamic code execution No
From automated source analysis of v0.1.0. βUsedβ means the capability is present in the source β more access means more to trust, not that itβs unsafe.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming β see below.
We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps β measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.
How agent discovery & health will work βAbout
Claude Workspace β Frontend Engineering Configuration
Production-grade Claude Code workspace for frontend teams. Drop into any React/TypeScript project β get expert-level AI pair programming with enforced standards, automated quality gates, and domain-specific knowledge.
Architecture
Why Claude Workspace?
Most developers copy-paste prompts. This repo gives you a version-controlled, team-shareable Claude Code configuration with:
- π― 18 domain-specific skills (React, TypeScript, a11y, SEO, Core Web Vitals, API, Auth, i18n)
- π€ 8 autonomous agents with auto-invocation rules
- β‘ 12 slash commands for full ticket-to-PR workflows
- π 10 hooks for deterministic quality gates
- π 5 path-scoped rules for contextual enforcement
- π¦ One-command install into any project
> Works with Claude Code β Anthropic's agentic CLI.
Quick Install
# One-command install into your project
curl -fsSL https://raw.githubusercontent.com/Piyush8296/claude-workspace/main/install.sh | bash -s -- /path/to/your/project
# Or manual copy
git clone https://github.com/Piyush8296/claude-workspace.git /tmp/cw
cp -r /tmp/cw/.claude /tmp/cw/CLAUDE.md /tmp/cw/.mcp.json your-project/
Then edit CLAUDE.md to match your stack and run claude.
What's Inside
claude-workspace/
βββ CLAUDE.md # Project memory (~60 lines, lean)
βββ install.sh # One-command installer
βββ .claude/
β βββ settings.json # Permissions, 10 hooks, env
β βββ agents/ (8)
β β βββ code-reviewer.md # Auto-invoked after code changes
β β βββ performance-auditor.md # Core Web Vitals & bundles
β β βββ accessibility-auditor.md # WCAG 2.1 AA compliance
β β βββ refactor-planner.md # Safe incremental migration
β β βββ doc-generator.md # Auto-writes JSDoc (background)
β β βββ architecture-reviewer.md # Module boundaries (background)
β β βββ test-writer.md # Writes tests (background)
β β βββ orchestrator.md # Unified health report (Opus)
β βββ commands/ (12)
β β βββ code-quality.md # /code-quality
β β βββ component-gen.md # /component-gen
β β βββ pr-review.md # /pr-review
β β βββ ticket.md # /ticket
β β βββ migrate.md # /migrate
β β βββ onboard.md # /onboard
β β βββ tdd.md # /tdd
β β βββ build-fix.md # /build-fix
β β βββ e2e.md # /e2e
β β βββ lighthouse.md # /lighthouse [url]
β β βββ perf-audit.md # /perf-audit [focus]
β β βββ deploy.md # /deploy (pre-deploy checklist)
β βββ rules/ (5)
β β βββ typescript.md # *.ts, *.tsx
β β βββ styling.md # *.css, tailwind.config.*
β β βββ testing.md # *.test.*, *.spec.*
β β βββ documentation.md # *.md, *.mdx
β β βββ env-config.md # .env*, config files
β βββ skills/ (18)
β β βββ react-patterns/SKILL.md # Component architecture
β β βββ state-management/SKILL.md # TanStack Query, Zustand, URL state
β β βββ testing-strategy/SKILL.md # TDD, factories, mocking
β β βββ css-architecture/SKILL.md # Tailwind, tokens, responsive
β β βββ performance-optimization/SKILL.md # Bundles, splitting, virtualization
β β βββ accessibility/SKILL.md # WCAG, ARIA, keyboard, focus
β β βββ systematic-debugging/SKILL.md # 4-phase root cause analysis
β β βββ frontend-design/SKILL.md # Aesthetics, typography, interactions
β β βββ design-intelligence/SKILL.md # Claude Design + Google Stitch
β β βββ security-audit/SKILL.md # XSS, CSP, secrets, auth, deps
β β βββ seo-fundamentals/SKILL.md # Meta tags, OG, SSR, crawlability
β β βββ structured-data/SKILL.md # JSON-LD schemas, GEO/AEO
β β βββ core-web-vitals/SKILL.md # LCP, INP, CLS fix recipes
β β βββ api-layer/SKILL.md # Typed fetch, interceptors, retry
β β βββ error-handling/SKILL.md # Error boundaries, normalization
β β βββ auth-patterns/SKILL.md # NextAuth, middleware, RBAC
β β βββ monorepo-patterns/SKILL.md # Turborepo, Nx, workspaces
β β βββ i18n/SKILL.md # next-intl, react-intl, RTL, plurals
β βββ hooks/scripts/
β βββ screenshot.sh # UI screenshot capture
βββ .mcp.json # MCP server config
βββ .gitignore
Skills (18)
| Skill | What It Covers | |-------|----------------| | react-patterns | Component architecture, composition, state handling order | | state-management | TanStack Query, Zustand, URL state, decision framework | | testing-strategy | TDD, factory pattern, mocking, coverage targets | | css-architecture | Tailwind patterns, design tokens, responsive, z-index | | performance-optimization | Bundle splitting, lazy loading, virtualization | | accessibility | WCAG 2.1 AA, ARIA, keyboard nav, focus management | | systematic-debugging | 4-phase root cause analysis | | frontend-design | Bold aesthetics, typography, micro-interactions | | design-intelligence | Claude Design + Google Stitch workflows | | security-audit | XSS, CSP, secrets, auth patterns, dependency vulns | | seo-fundamentals | Meta tags, OG/Twitter cards, SSR vs CSR, crawlability | | structured-data | JSON-LD schemas (Product, Article, FAQ, Breadcrumb), GEO/AEO | | core-web-vitals | LCP, INP, CLS β diagnosis + specific React/Next.js fix recipes | | api-layer | Typed fetch clients, interceptors, retry with backoff, AbortController | | error-handling | Error boundaries hierarchy, custom error classes, normalization | | auth-patterns | NextAuth/Auth.js, middleware protection, RBAC, token refresh | | monorepo-patterns | Turborepo/Nx pipelines, shared packages, changesets | | i18n | next-intl/react-intl, locale routing, RTL, ICU pluralization |
Agents (8)
| Agent | Model | Trigger | |-------|-------|---------| | code-reviewer | Sonnet | Proactively after any code change | | performance-auditor | Sonnet | Performance complaints or pre-launch | | accessibility-auditor | Sonnet | UI work or a11y issues reported | | refactor-planner | Sonnet | Before restructuring or migration | | doc-generator | Haiku | After code changes (background) | | architecture-reviewer | Sonnet | Adding modules or restructuring (background) | | test-writer | Sonnet | After features or bug fixes (background) | | orchestrator | Opus | Full project health check on demand |
Commands (12)
| Command | What it does | |---------|-------| | /code-quality | Lint + typecheck + manual review | | /component-gen | Scaffold component + test + barrel | | /pr-review | Review current branch changes | | /ticket | Full ticket-to-PR workflow | | /migrate | Guided migration with safety checks | | /onboard | Explore and document the codebase | | /tdd | Red-green-refactor TDD cycle | | /build-fix | Auto-diagnose and fix build errors | | /e2e | Generate Playwright E2E tests | | /lighthouse [url] | Run Lighthouse audit, parse scores, generate fix list | | /perf-audit [focus] | Bundle analysis, dep weight, images, fonts, third-party | | /deploy | Pre-deployment readiness checklist |
Rules (5)
| Rule | Glob Pattern | What It Enforces | |------|-------------|------------------| | typescript | *.ts, *.tsx | Strict types, no any, interface conventions | | styling | *.css, tailwind.config.* | Design tokens, utility-first, responsive | | testing | *.test.*, *.spec.* | AAA pattern, factories, coverage thresholds | | documentation | *.md, *.mdx | Structure, frontmatter, linking | | env-config | .env*, config.ts | Zod validation, typed config, no hardcoded secrets |
Hooks (10)
| Hook | Event | What It Does | |------|-------|--------------| | Branch guard | PreToolUse | Blocks edits on main | | File protection | PreToolUse | Blocks .env, production configs, secrets | | Prettier | PostToolUse | Auto-formats JS/TS/JSX/TSX | | TypeScript check | PostToolUse | tsc --noEmit on TS changes | | Test runner | PostToolUse | Runs related tests | | Dep installer | PostToolUse | Auto-install on package.json change | | Auto-commit | PostToolUse | Stages + commits every edit | | UI screenshot | PostToolUse | Captures app on UI file changes | | Notification | Stop | macOS notification when done | | Vercel preview | Stop | Deploy preview after clean build |
Design Tools
- Claude Design β Visual prototyping from conversation, exports to React
- Google Stitch β Free multi-screen UI generation, exports to Tailwind/Vue/Flutter/SwiftUI
Extending
- Add skills:
.claude/skills//SKILL.md - Add agents:
.claude/agents/.md - Add rules:
.claude/rules/.md(withglobs:frontmatter) - Add commands:
.claude/commands/.md - Update
CLAUDE.mdQuick Facts for your stack
Philosophy
- Lean CLAUDE.md β under 65 lines; detail lives in skills and rules
- Progressive disclosure β Claude loads only what the task needs
- Deterministic gates β hooks enforce formatting and checks regardless of model
- Composable β skills cross-reference, agents delegate, commands orchestrate
License
MIT
Source & license
This open-source MCP server is cataloged on AgentStack and links to its original source β we do not rehost the code.
- Author: Piyush8296
- Source: Piyush8296/claude-workspace
- License: MIT
- Homepage: https://www.linkedin.com/in/piyush-sde
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet β be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.