Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@LaunchFrame MCP scaffold a new NestJS module for 'orders' following project conventions"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
MCP server that gives AI agents (Claude Code, Cursor, etc.) architectural knowledge about your LaunchFrame project — auth patterns, entity conventions, queue usage, cron jobs, webhooks, and more.
Running launchframe init automatically configures this MCP server in your project. No manual setup needed.
Related MCP server: Laravel AI MCP Server
Instead of bloating CLAUDE.md with rules, agents call tools on demand to pull the exact pattern they need:
Tool | Description |
| Auth system, guard hierarchy, session flow |
| Exact decorator for a given auth need |
| Guard class + import for admin/user/credits |
| How feature gates are stored and queried |
| Copy-paste feature check snippet |
| Decorator + guard for route-level credit deduction |
| Code snippet for programmatic credit addition |
| Overview of free/subscription/credits/hybrid modes |
| Available Bull queues and their purpose |
| Producer injection snippet for a given queue |
| Processor class scaffold for a queue + job name |
| Receipt/processing separation, retry logic |
| Controller + processor scaffold for a provider/event |
| Where crons live, available schedules, rules |
| Cron method scaffold to add to CronService |
| NestJS module folder structure and conventions |
| Full module scaffold (module + service + controller + entity) |
| TypeORM entity conventions and required decorators |
| TypeORM entity file scaffold |
| Centralized |
| Base / Multi-tenant / B2B2C variant differences |
| Full DB schema with tables, columns, types, and common SQL snippets |
| Subscription plans system: plan groups, annual billing, API shape |
| Overall architecture: services, modules, patterns, infrastructure |
| Email system: sending patterns, templates, built-in templates |
| Handlebars template stub for a new transactional email |
| NestJS code to send an email (direct or queued) |
| Start Docker services (detached) |
| Stop all running Docker services |
| Build Docker images (all or specific service) |
| Fetch a snapshot of Docker service logs |
| Destroy ALL Docker resources — irreversible |
| Run all pending TypeORM migrations |
| Create a new empty TypeORM migration file |
| Revert the most recently applied migration |
| Execute a SQL query (local or remote via SSH) |
| List available optional services |
| Install an optional service non-interactively |
| List available optional modules |
| Install a module non-interactively |
| Build production images and push to GHCR |
| Deploy latest images to VPS via SSH |
| Sync plan features to production DB — destructive |
| Start the waitlist service locally |
| Stop the local waitlist service |
| Build and deploy waitlist service to VPS |
| Fetch waitlist service logs from VPS |
| Create a random test user in the local DB |
| Run npm install inside a service via Docker |
| Generate and inject logo/favicon assets |
| Show local service cache info |
| Force-update the local service cache |
| Delete the entire local service cache |
If you prefer a global install:
npm install -g @launchframe/mcpThen in .mcp.json:
{
"mcpServers": {
"launchframe": {
"command": "launchframe-mcp"
}
}
}Get the overall architecture of the LaunchFrame project: services, backend module layout, key patterns, frontend structure, and infrastructure.
Inputs: no named parameters.
Get the exact decorator and import for a specific auth need.
Inputs: need (required).
Get the guard class, import path, and decorator combo for a specific guard type.
Inputs: guard (required).
Get full auth system overview: guard hierarchy, session flow, Better Auth setup, roles, and decorator system.
Inputs: no named parameters.
Delete the entire local LaunchFrame service cache. Services will be re-downloaded on next use.
Inputs: no named parameters.
Show information about the local LaunchFrame service cache (location, size, cached services, last update time).
Inputs: no named parameters.
Force-update the local LaunchFrame service cache by re-downloading cached services.
Inputs: no named parameters.
Execute a SQL query and return the results as text.
Workflow:
Call database_schema first to learn table/column names.
Call this tool with a complete, valid SQL statement.
Read the returned text — it is raw psql output (column headers + rows).
Local (default): runs against the Docker Compose database on the developer machine.
→ Requires launchframe docker:up to be running.
Remote (remote=true): SSHs into the VPS and runs against the production database. → Will ask the user for confirmation before executing. → Only use when the user explicitly asks about production/live data.
Supported statements: SELECT, INSERT, UPDATE, DELETE, EXPLAIN, etc. Always end the SQL with a semicolon. Quote identifiers that are reserved words (e.g. "user", "order").
Inputs: projectPath (required), sql (required), remote.
Build production Docker images and push to GitHub Container Registry. Optionally build a specific service only.
Inputs: projectPath (required), service.
Sync subscription plan features from the local database to the production database. DESTRUCTIVE: truncates subscription_plan_features on production. Will prompt for confirmation before proceeding.
Inputs: projectPath (required).
Deploy the latest images to the VPS and restart all production services via SSH.
Inputs: projectPath (required).
Create a random test user in the local database. Generates a unique email + bcrypt password hash and inserts via docker exec. Creates a demo project if the project uses multi-tenancy.
Inputs: projectPath (required).
Generate and inject logo/favicon assets across all frontend services from an SVG file. Defaults to /logo.svg if no path is given.
Inputs: projectPath (required), svgPath.
Run npm install inside a service directory using a node:20-alpine Docker container (matches the build environment). Use this to add packages or update package-lock.json.
Inputs: projectPath (required), service (required), packages.
Build Docker images for a LaunchFrame project (all services or a specific one).
Inputs: projectPath (required), service.
Destroy ALL Docker resources for a LaunchFrame project (containers, volumes, images, network). IRREVERSIBLE — all local data including database volumes will be lost. Will prompt for confirmation before proceeding.
Inputs: projectPath (required).
Stop all running Docker services for a LaunchFrame project.
Inputs: projectPath (required).
Fetch a snapshot of Docker service logs (non-streaming). Returns the last N lines.
Inputs: projectPath (required), service, tail.
Start Docker services for a LaunchFrame project. Always runs detached (background). Use cli_docker_logs to inspect output afterward.
Inputs: projectPath (required), service.
Create a new empty TypeORM migration file with the given name.
Inputs: projectPath (required), name (required).
Revert the most recently applied TypeORM database migration.
Inputs: projectPath (required).
Run all pending TypeORM database migrations against the local database.
Inputs: projectPath (required).
Install a module into a LaunchFrame project. Runs non-interactively (skips confirmation). Rebuilds affected containers and restarts the stack in detached mode.
Inputs: projectPath (required), name (required).
List all available modules that can be added to a LaunchFrame project (e.g., feature-flags, multi-tenancy).
Inputs: projectPath (required).
Install an optional service into a LaunchFrame project. Runs non-interactively (skips prompts). Env vars will be empty — configure them manually in infrastructure/.env afterward.
Inputs: projectPath (required), name (required).
List all available optional services that can be added to a LaunchFrame project (e.g., waitlist, docs, customers-portal).
Inputs: projectPath (required).
Build and deploy the waitlist service to the VPS via SSH.
Inputs: projectPath (required).
Stop the locally running waitlist service.
Inputs: projectPath (required).
Fetch a snapshot of waitlist service logs from the VPS (non-streaming). Returns the last N lines.
Inputs: projectPath (required), tail.
Start the waitlist service locally using Docker Compose.
Inputs: projectPath (required).
Get the code snippet for programmatically adding credits to a user with a specific transaction type.
Inputs: transactionType (required).
Get the decorator + guard pattern for deducting credits on a route.
Inputs: no named parameters.
Get an overview of all monetization strategies (free, subscription, credits, hybrid) and when to use each.
Inputs: no named parameters.
Get the LaunchFrame cron job pattern: where jobs live, available CronExpression presets, and module registration rules.
Inputs: no named parameters.
Scaffold a new cron method to add to CronService (src/jobs/cron.service.ts).
Inputs: methodName (required), schedule (required), queueName.
Returns the full database schema (all tables, columns, types, relations) plus ready-made SQL snippets for common questions like user counts, active sessions, subscription plans, credit balances, etc. Call this before running cli_database_query when the user asks a data question.
Inputs: no named parameters.
Get an overview of the LaunchFrame email system: sending patterns (direct vs queue-based), template conventions, built-in templates, and environment setup.
Inputs: no named parameters.
Generate NestJS code to send a transactional email — either by injecting MailService directly (direct) or via the emails Bull queue (queued).
Inputs: templateName (required), method (required), contextVars (required).
Generate a Handlebars (.hbs) template stub for a new LaunchFrame transactional email, following project conventions.
Inputs: templateName (required), contextVars (required).
Get TypeORM entity conventions for LaunchFrame: required decorators, naming strategy, column types, relations, and multi-tenancy.
Inputs: no named parameters.
Generate a TypeORM entity file following LaunchFrame conventions.
Inputs: entityName (required), tableName (required), primaryKeyType, multiTenant, +1 more.
Get environment variable conventions for LaunchFrame: single centralized .env location, variable naming rules, full key variable reference, and how to add new variables.
Inputs: no named parameters.
Get a copy-paste TypeScript snippet for checking a feature gate by code and type.
Inputs: featureCode (required), featureType (required).
Get the feature gate system overview: how features are stored, how to query them, and the check pattern.
Inputs: no named parameters.
Get the NestJS module folder structure, conventions, and rules used in LaunchFrame.
Inputs: no named parameters.
Generate a NestJS module scaffold (module + service + optional controller + optional entity) following LaunchFrame conventions.
Inputs: moduleName (required), withController, withEntity.
List all available Bull queues in LaunchFrame with their purpose and usage rules.
Inputs: no named parameters.
Get a Bull processor class scaffold for a given queue and job name.
Inputs: queueName (required), jobName (required).
Get the code to inject and use a Bull queue as a producer in a NestJS service.
Inputs: queueName (required).
Get an overview of the subscription plans system: plan groups, annual billing toggle, API response shape, and key files.
Inputs: no named parameters.
Get an overview of LaunchFrame project variants: Base (B2B single-tenant), Multi-tenant, and B2B2C. Explains how to read the active variant from the .launchframe file and what behavioral differences to account for when writing new code.
Inputs: no named parameters.
Get an overview of the LaunchFrame webhook architecture: receipt/processing separation, WebhookLog entity, Bull queue, and retry cron.
Inputs: no named parameters.
Get a scaffold for a new webhook handler: controller receipt + Bull processor for a given provider and event type.
Inputs: provider (required), eventType (required).
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
Scored across 53 tools
Most tools have distinct purposes, but some overlap exists, such as between cli_cache_clear, cli_cache_update, and cli_database_query (which involves cache considerations). However, descriptions clarify boundaries, and tools are generally well-separated by domain (e.g., auth, cli, database).
Tool names follow a highly consistent pattern: most use a clear domain_prefix_verb_noun structure (e.g., auth_get_overview, cli_cache_clear, database_schema). This predictability makes it easy for an agent to understand the tool's scope and action.
With 53 tools, the count is excessive for a single server, making it overwhelming and difficult to navigate. This many tools suggests the server is trying to cover too many domains (e.g., CLI operations, database, auth, email) without proper scoping, which can lead to confusion and inefficiency.
The tool set provides comprehensive coverage for the LaunchFrame project domain, including CRUD-like operations, scaffolding, system overviews, and lifecycle management across various components (e.g., auth, database, CLI, email). There are no obvious gaps, and tools support end-to-end workflows effectively.
Project memory, documentation and controlled secret access for AI coding agents.
Serves your design system and coding standards to coding agents, so they stop guessing.
Shared memory for coding agents. Stop re-explaining your codebase every session.
Structured knowledge base for AI agent solutions. Search, explore, and retrieve build logs.
