Chat with your code

Introducing Monty: your new VS Code assistant

What is Monty?

A chat-style assistant, inside your IDE

Monty combines the intelligence of LLMs with the context of your code using a technique called RAG (retrieval augmented generation). Monty will automatically update its context to find code relevant to the task at hand.

  • Faster onboarding: Come up to speed on a new codebase with documentation and explanations generated on-the-fly
  • Better planning: Monty can evaluate or propose an implementation plan for bug fixes or new features, and can even suggest code

Bring your own key

Try Monty for free. After the trail, Monty is free for personal use with your own LLM API key. Currently Monty works with OpenAI, Azure OpenAI, Anthropic, and Google AI Studio, as well as OpenAI API-compatible providers such as Groq and Ollama.

Monty learns your code by building a local index

This allows Monty to automatically find relevant code

Privacy Friendly

The index is built locally and never leaves your machine. Code is only seen by your LLM.

Real-time Indexing

Codebases are constantly changing. Monty keeps its code index up to date so its knowledge is never stale.

Language Agnostic

As long as your code is in text files, Monty can read it. Doesn't matter what language you use.

Add to code context

Now with manual context control

Monty provides helpers that let you precisely specify the relevant files, functions, classes, methods, and code blocks.

Definition or selection.
Add this function/class definition or highlighted text
File.
Add this entire file
Callers.
Add the functions and methods that call this
Callees.
Add the functions and methods that this calls
Neighborhood.
Add the callers and callees of this
2-hop neighborhood.
Add the neighbors of this and the neighbors' neighbors

Try Document Mode

Monty can optionally use a markdown document instead of chat history when you’re trying to produce a shareable document, ticket, or specification.

App screenshot
Curate Instructions:
Ensure repeatability, efficiency, and quality from your LLM.
Use Templates:
Minimize rework and ambiguity in your documents.
Edit Collaboratively:
Jointly edit with Monty.

In this mode, Monty uses the content of your document and instruction to search for relevant code. As you update the document, Monty’s context updates as well.

Monty is free for personal projects

Interested in a commercial license? Drop us a line

Frequently asked questions

What are the system requirements?

Monty requires either Windows (x86_64), Linux (x86_64), or macOS (Apple Silicon) with VSCode and Git installed.

Which IDEs does Monty support?

At this time only VSCode is supported, though we plan to support other IDEs in the future (e.g. JetBrains).

Does Monty work with private repositories?

Monty works with local repositories and builds local indices. The only code that leaves your system is what's sent to your LLM.

Does Monty work with remote workspaces?

VSCode remote workspaces and WSL (Windows Subsystem for Linux) are both supported.

Which LLMs are supported?

Monty currently supports several LLM providers, including OpenAI and OpenAI compatible providers such as Groq and Ollama, as well as Azure OpenAI, Anthropic, and Google AI Studio.