Fastapi Hexagon Generator Mcp — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited Fastapi Hexagon Generator Mcp (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
An MCP (Model Context Protocol) server that lets Claude — and any other MCP-compatible client — scaffold production-ready FastAPI projects following hexagonal (ports & adapters) architecture.
It generates the full module: domain entities, ORM models, Pydantic schemas, repositories, use cases, and API routes — with the wiring between them, validation, search, filters, and relationships all handled for you.
Hexagonal architecture is great for separation of concerns, but the boilerplate is painful: every new entity touches 6+ files across 4 layers. This MCP automates the mechanical parts (file generation, field propagation, repository mapping, dependency wiring) while leaving the parts that need real judgment (domain rules, custom validations) as TODOs you complete with the LLM.
The result: ask Claude for "a Product CRUD with name, price, and category" and get a working, well-structured FastAPI module in seconds.
generate_builtin.The server runs via uvx, so you don't need to install anything globally — just configure your MCP client to invoke it.
Add to your claude_desktop_config.json or ~/.claude.json:
{
"mcpServers": {
"hexagonal-generator": {
"type": "stdio",
"command": "uvx",
"args": [
"--from",
"fastapi-hexagon-mcp",
"hexagonal-mcp"
],
"env": {}
}
}
}Pin a specific version with fastapi-hexagon-mcp==X.Y.Z if you need reproducibility.
pip install fastapi-hexagon-mcp
hexagonal-mcp # runs the server over stdioOnce configured, prompt your LLM in natural language. Example:
Generate a CRUD module called Product using the hexagonal-generator MCP. Define its fields:name(str, max 100, searchable),price(float, required),description(str, max 500, nullable), andcategory(str, max 50, searchable). Add a validation:namemust start with"PRODUCT: ". Then generate a module Store withname(str, max 100, searchable),address(str, required), andproducts(foreign key to Product). Wire the modules and complete all remaining TODOs.
The model will orchestrate the MCP tools below to produce a working project.
| Tool | What it does |
|---|---|
generate_builtin | Bootstrap a project skeleton (auth, base infra, alembic). |
generate_crud | Scaffold a new CRUD module's files across all layers. |
define_fields | Add fields to a module and propagate them to entities, models, schemas, repositories, and filters. |
add_relationship | Add a foreign-key relationship between two modules. |
wire_module | Register a module's dependencies (repository + use cases) in the DI container and routes. |
list_todos | List remaining TODOs in a module so the LLM can complete them. |
complete_todos | Help complete TODOs that need judgment (custom validation, business rules). |
generate_builtin — create the project skeleton.generate_crud — scaffold a module.define_fields — fill in the fields.add_relationship (optional) — connect modules.wire_module — plug it into the app.list_todos / complete_todos — finish the parts that need understanding.src/
└── <module>/
├── domain/
│ ├── entities.py # dataclasses
│ └── repositories.py # repository protocol
├── application/
│ ├── schemas.py # Pydantic request/response models
│ └── use_cases.py # business logic
├── infrastructure/
│ ├── models.py # SQLAlchemy ORM
│ └── database.py # repository implementation
└── routes.py # FastAPI routeruvx (or pip + a venv) to run the serverMIT — see LICENSE.
~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.