baa
Supports generating Electron starter projects and structurally verifying generated manifests and required entry files.
Supports generating Flutter starter projects and structurally verifying generated manifests and required entry files.
Supports generating Jetpack Compose starter projects and structurally verifying generated manifests and required entry files.
Supports generating Next.js starter projects and structurally verifying generated manifests and required entry files.
Supports generating React starter projects and structurally verifying generated manifests and required entry files.
Supports generating Roblox starter projects and structurally verifying generated manifests and required entry files.
Supports generating Tauri starter projects and structurally verifying generated manifests and required entry files.
Supports generating Unity starter projects and structurally verifying generated manifests and required entry files.
This server does not currently have a deployable release on Glama. Follow the repository's setup instructions to install and configure it for your MCP client, including any required credentials and local dependencies.
⚡ BUILD ANYTHING
The agent-native engineering workbench
Understand systems. Generate projects. Inspect evidence. Verify what changed.
Get started · What it does · Use with agents · Developer guide · Safety
What is BAA?
Build Anything (BAA) is a local-first toolkit for developers and AI agents. It brings project workflows and software-inspection capabilities together behind a command-line interface and a Model Context Protocol (MCP) server.
BAA is designed to help an agent or developer move from understanding a codebase to creating a project and checking the resulting structure—while keeping the work inspectable and the limits of each check explicit.
The BAA principle: make useful engineering actions available as bounded tools, return evidence, and never claim a check happened when it did not.
Related MCP server: Agentic MCP Server
✨ What BAA does
Capability | What it is for |
Project discovery | Inspect a local source tree and produce a bounded analysis of its structure and candidates. |
Project generation | Create a starter project for a supported target without silently overwriting existing files. |
Structural verification | Check generated manifests and required entry files. |
Source and runtime inspection | Use the repository's existing analysis and evidence tools to investigate supported software artifacts. |
Agent integration | Expose tools through MCP so compatible AI clients can use BAA in their workflows. |
CLI workflows | Run setup, analysis, generation and verification from a terminal. |
BAA currently includes starter targets for React, Next.js, Vue, Flutter, React Native, SwiftUI, Jetpack Compose, Electron, Tauri, Unity, Unreal and Roblox. See the project documentation for the precise capabilities and requirements of each target.
🚀 Get started
Requirements
Node.js 22.23.4+, 24.22.1+, or 26.12.1+
npm 11+
Install from a source checkout
git clone https://github.com/devMarksKnobb/BAA.git
cd BAA
npm ci
npm run build:cached
node scripts/baa.mjs setupIf your local checkout directory has a different name, run the commands from that directory. Build BAA before starting its MCP server for the first time.
Try the project workflow
# See available starter targets
node scripts/baa.mjs project targets
# Analyse an existing local project
node scripts/baa.mjs project analyze ./existing-project --out ./baa-analysis.json
# Generate a starter project
node scripts/baa.mjs project generate --target react --out ./new-project --name "My App"
# Check the generated project structure
node scripts/baa.mjs project verify ./new-projectUse baa --help (or node scripts/baa.mjs --help from a checkout) for the current command reference.
🤖 Connect an MCP client
Add a BAA server entry to the MCP client's configuration. Replace the example paths with absolute paths on your machine:
{
"mcpServers": {
"baa": {
"command": "node",
"args": ["/absolute/path/to/BAA/scripts/baa.mjs", "mcp"],
"cwd": "/absolute/path/to/BAA"
}
}
}BAA supports the repository's configured MCP transports, including stdio and the documented Streamable HTTP mode. Consult the installation and MCP documentation for client-specific setup and transport details.
Core project tools
Tool | Purpose |
| List supported project-generation targets. |
| Inventory a local source tree and extract deterministic candidates. |
| Generate a target-specific starter without overwriting existing files. |
| Check a generated manifest and required entry files. |
These project tools sit alongside BAA's broader source and runtime evidence capabilities.
🧭 Navigate the repository
BAA/
├── .agents/ Agent instructions and reusable workflows
├── .github/ GitHub automation and contribution templates
├── assets/ Project artwork and supporting assets
├── bridge/ External-tool and platform bridges
├── docs/ Guides, architecture, references and verification records
├── examples/ Example configurations and workflows
├── native/ Native integration components
├── scripts/ CLI entry points, build and verification helpers
├── src/ Application source and domain modules
├── tests/ Automated tests and fixtures
├── third_party/ Vendored and referenced third-party projects
├── README.md Project overview and first steps
├── CONTRIBUTING.md
├── SECURITY.md
└── package.jsonFor a deeper explanation, start with docs/BAA-QUICKSTART.md, then browse docs/index.md and docs/architecture.mermaid.
🛠️ Work on BAA
npm ci
npm run build:cached
npm run test:baa
npm run test:localThe focused BAA test command covers project analysis, generation and structural verification. The full suite and integrations may require additional platform tools and fixtures; read docs/testing.md and docs/installation.md before running broader checks.
Useful project references:
🛡️ Trust, safety and honest verification
BAA may inspect source projects, archives, binaries and generated files. Treat all such inputs—including model output—as untrusted.
Keep generated writes within explicit output directories.
Review proposed changes before using them in important projects.
Do not place secrets, credentials or private keys in source fixtures or issue reports.
Respect licenses and terms for third-party components.
Distinguish structural checks from compilation, execution, accessibility testing and visual validation.
A passing structural check is not proof that an application is production-ready. BAA should report what it actually checked, not imply that unrun checks succeeded.
🌍 Documentation languages
Localized introductions are available in the repository root. Choose the language that works best for you; the canonical English guide remains the reference for current commands and setup details.
Language | README |
العربية | |
Deutsch | |
Ελληνικά | |
English | |
Español | |
فارسی | |
Français | |
Bahasa Indonesia | |
Italiano | |
日本語 | |
한국어 | |
മലയാളം | |
Polski | |
Português (Brasil) | |
Română | |
Русский | |
ไทย | |
Türkçe | |
Українська | |
اردو | |
Tiếng Việt | |
简体中文 | |
繁體中文 |
Localized files provide translated introductions; links and command examples should be checked against the canonical guide when updating releases.
License
BAA is released under the MIT License. See LICENSE and THIRD_PARTY_NOTICES.md for details and third-party terms.
This server cannot be deployed
Maintenance
Related MCP Connectors
Code intelligence platform for AI agents. 20 tools for architecture, security & impact analysis.
AI-native programming language and toolchain for accountable, verifiable agent-driven development.
Cross-agent artifact workspace with provenance across Claude Code, Codex, Cursor, LangGraph.
Context and control for any AI. Agent identity, scoped access, and proof across tools.
Related MCP Servers
- AlicenseNot gradedqualityDmaintenanceEnables AI agents to detect development environments, install missing tools, scan local code projects, and generate visual reports.3 npmMIT
- AlicenseAqualityDmaintenanceActs as the 'Hands and Eyes' for an Autonomous AI Agent, bridging Large Language Models and your local development environment to enable safe file manipulation, context reading, command execution, and documentation verification.92ISC
- AlicenseNot gradedqualityAmaintenanceEnables AI clients to safely read, search, understand, and edit local project code and files, with Git inspection, code indexing, and controlled command execution within permissioned workspaces.4 npm6MIT
- AlicenseNot gradedqualityBmaintenanceEnables AI assistants to securely inspect, modify, and manage local projects with trusted filesystem access, safe Git operations, controlled task execution, and developer runtime process management over stdio or HTTP.4Apache 2.0