Tidewave Rails
Better agentic Rails development, runtime-level tools for your agent to talk to your running app.
README
Tidewave
Tidewave is the coding agent for full-stack web app development. Integrate Claude Code, OpenAI Codex, and other agents with your web app and web framework at every layer, from UI to database. See our website for more information.
This project can also be used as a standalone Model Context Protocol server.
Installation
You can install Tidewave by running:
bundle add tidewave --group development
or by manully adding the tidewave gem to the development group in your Gemfile:
gem "tidewave", group: :development
Now make sure Tidewave is installed and you are ready to connect Tidewave to your app.
Troubleshooting
Using multiple hosts/subdomains
If you are using multiple hosts/subdomains during development, you must use *.localhost, as such domains are considered secure by browsers. Additionally, add the following to config/initializers/development.rb:
config.session_store :cookie_store,
key: "__your_app_session",
same_site: :none,
secure: true,
assume_ssl: true
And make sure you are using rack-session version 2.1.0 or later.
The above will allow your application to run embedded within Tidewave across multiple subdomains, as long as it is using a secure context (such as admin.localhost, www.foobar.localhost, etc).
Content security policy
If you have enabled Content-Security-Policy, Tidewave will automatically enable "unsafe-eval" under script-src in order for contextual browser testing to work correctly. It also disables the frame-ancestors directive.
Production Environment
Tidewave is a powerful tool that can help you develop your web application faster and more efficiently. However, it is important to note that Tidewave is not meant to be used in a production environment.
Tidewave will raise an error if it is used in any environment where code reloading is disabled (which typically includes production).
Configuration
You may configure tidewave using the following syntax:
config.tidewave.team = { id: "my-company" }
The following config is available:
-
allow_remote_access- Tidewave only allows requests from localhost by default, even if your server listens on other interfaces, for security purposes. Read our security guidelines for more information and when to allow remote access (if you know what you are doing) -
logger_middleware- The logger middleware Tidewave should wrap to silence its own logs -
preferred_orm- which ORM to use, either:active_record(default) or:sequel -
team- set your Tidewave Team configuration, such asconfig.tidewave.team = { id: "my-company" }
Available tools
-
execute_sql_query- executes a SQL query within your application database, useful for the agent to verify the result of an action -
get_docs- get the documentation for a given module/class/method. It consults the exact versions used by the project, ensuring you always get correct information -
get_logs- reads logs written by the server -
get_models- lists all modules in the application and their location for quick discovery -
get_source_location- get the source location for a given module/class/method, so an agent can directly read the source skipping search -
project_eval- evaluates code within the Rails application itself, giving the agent access to your runtime, dependencies, and in-memory data
[!NOTE]
Why no tools for routes, associations, etc?
Tidewave does not include tools for listing your routes, associations, etc. because agents are better off reading their respective source files, which gives agents more context and enable them to perform any necessary edit without additional tools calls.
Instead, Tidewave aims to fill in missing gaps, such as evaluating code inside your Rails app (without starting new instances) and finding source location, which can be tricky, even with grepping, due to meta-programming and the different places Bundler can install your dependencies.
Acknowledgements
A thank you to Yorick Jacquin for the initial version of this project.
Development
Run the Minitest suite with:
bundle exec ruby -Itest test/all_test.rb
License
Copyright (c) 2025 Dashbit
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.