VibemasterMCP

The #1 MCP orchestrator that makes Claude think like a senior engineer — 58 tools, persistent memory, GSD planning, RAG search, auto-review, quality scoring, and security hardening.

Quick Start

Install

npm i vibemaster-mcp

The interactive setup wizard launches automatically after install. Choose your runtimes and you're done.

Manual Setup

npx vibemaster-mcp-setup              # Interactive wizard
npx vibemaster-mcp-setup --all        # Auto-configure everything
npx vibemaster-mcp-setup claude cursor # Specific runtimes
npx vibemaster-mcp-setup --uninstall   # Remove all configs

Supported Runtimes

| Runtime | Config Written To | |---------|------------------| | Claude Code | claude mcp add or .claude.json | | OpenCode | opencode.json | | Codex (OpenAI) | .codex/config.json | | Gemini CLI | .gemini/settings.json | | Cursor | AppData/Cursor/mcp.json | | Windsurf | AppData/Windsurf/mcp.json | | VS Code | settings.json (mcp.servers) | | Cline | cline_mcp_settings.json | | Augment | settings.json (mcp.servers) | | GitHub Copilot | .vscode/mcp.json | | Trae | settings.json | | Kilo, Qwen Code, CodeBuddy, Antigravity | Config files |

Verify

Open your AI tool and run:

health_check

First Steps

bootstrap              Set up environment + install MCPs
health_check           Verify everything works
remember               Save your first memory
plan "build a REST API" See VibeMaster plan a project

What It Does

Orchestration

• orchestrate — Route any query to the right MCP combination
• plan — Generate execution plan before coding
• autopilot — End-to-end task with self-correction
• agent_team — Decompose large tasks into specialist agents

Memory & Context

• remember / recall / forget — Persistent knowledge across sessions
• memory_summary — View all stored memories
• smart_context — Combine docs + memory + project info
• code_context — Get relevant code for your current task

Code Intelligence

• debug_session — Deep static analysis (60 rules)
• code_review — PR-style review with 35+ rules
• quality_report — Score code 0-100
• auto_patch — Apply safe mechanical fixes
• agentic_fix — Full analyze → patch → verify cycle
• complexity_check — Track complexity over time
• call_graph — Build call graph + impact analysis
• diff_analyze — Show bugs introduced by your change
• dataflow_analyze — Trace tainted data to dangerous sinks

RAG & Search

• index_codebase — Build semantic search index with vector embeddings
• code_search — Search codebase by meaning, not text
• explain_codebase — Generate architecture documentation

GSD Project Management

• gsd_new — Start project with clarifying questions
• gsd_plan — Generate full project roadmap
• gsd_execute — Execute a phase or milestone
• gsd_uat — Run UAT verification
• gsd_status — Track project progress

Security & Quality

• secret_audit — Scan configs for exposed API keys
• security_report — Full security posture report
• trust_report — Score MCP trustworthiness
• health_check — Check which MCPs are healthy

Environment Setup

• bootstrap — One-command environment setup
• detect_stack — Identify your tech stack
• install_mcps — Install missing MCP servers
• discover_mcps — Search 4,000+ MCP registry
• generate_claude_md — Generate CLAUDE.md system prompt

Tools Reference

| Tool | Description | |------|-------------| | orchestrate | Intelligently orchestrate multiple MCPs | | debug_session | Full bug analysis with pipeline modes | | auto_patch | Apply safe mechanical fixes | | agentic_fix | Analyze → patch → verify cycle | | code_review | PR-style review with static analysis | | quality_report | Code quality score 0-100 | | diff_analyze | Show bugs introduced by changes | | dataflow_analyze | Taint flow analysis | | complexity_check | Complexity scoring + trends | | call_graph | Function call relationships | | refactor_suggest | Concrete refactoring ideas | | analyze_project | Multi-file architecture health | | generate_tests | Vitest regression test generation | | analyze_prompt | AI prompt quality scoring | | run_workflow | Multi-step analysis workflow | | remember / recall / forget | Persistent memory | | plan / autopilot | Planning & autonomous execution | | agent_team | Task decomposition into agents | | gsd_new / gsd_plan / gsd_execute | GSD project management | | gsd_uat / gsd_status / gsd_list | UAT & tracking | | index_codebase / code_search | RAG semantic search | | bootstrap / install_mcps | Environment setup | | discover_mcps | Search 4,000+ MCPs | | health_check / metrics | Monitoring | | secret_audit / security_report | Security | | smart_context / code_context | Intelligent context | | session_report / daily_standup | Session analytics |

Workflows

| Workflow | Description | |----------|-------------| | full_review | Complete analysis: bugs, quality, complexity, tests | | security_audit | Taint flow + injection vulnerability checks | | pre_commit | Fast check before committing | | onboarding | First-time analysis of a new file | | fix_and_verify | Apply fixes, generate tests, verify |

Skills

Built-in structured workflows with automatic planning:

• implement_feature — Plan and implement a new feature
• fix_bug — Diagnose and fix a bug
• security_hardening — Harden code against attacks
• release_ready — Prepare for production release
• write_tests — Generate comprehensive tests
• code_review — Full code review workflow
• explain_codebase — Architecture documentation

run_skill skill_id=fix_bug query="login returns 500 on empty email"

Orchestrated MCPs

VibeMaster discovers and orchestrates these MCPs:

Tier 1 — Core

• context7 — Up-to-date library docs
• sequential-thinking — Structured reasoning
• playwright — Browser automation
• taskmaster — Task management
• desktop-commander — Shell commands

Tier 2 — Dev Workflow

• github — Issues, PRs, repos
• filesystem — File read/write
• brave-search — Live web search
• markitdown — PDF/Office conversion
• gitguardian — Secret scanning

Tier 3 — Remote Services

• supabase, sentry, stripe, vercel, linear

Requirements

• Node.js >= 20.0.0
• npm >= 9.0.0

Development

npm install
npm run build
npm test
npm run start

Version History

| Version | Key Addition | |---------|-------------| | v8.0.0 | 58 tools, GSD, RAG, agent teams, sync | | v8.0.1 | Setup wizard, better-sqlite3 v12, cross-platform fixes |

License: MIT © SnaiilyDevelopment

GitHub: https://github.com/SnaiilyDevelopment/VibemasterMCP