Empryo

Empryo

MCP.Pizza Chef: proxysoul

This is a full coding assistant application for Mac, Linux, and Windows that reads your entire project first, drawing a live map of how every file and function connects before it changes anything. That means it can warn you what else might be affected by an edit before making it, and it makes precise changes instead of guessing with plain text search-and-replace. You install it with a short terminal command or a package manager, then connect your own account from a provider like Anthropic, OpenAI, or Google — or skip that entirely and run a free, private local model instead.

Coding

Use This MCP client To

Fix a bug without breaking other parts of my app See what else might break before I change a piece of code Update the same kind of change across many files at once Undo any step and rewind my code back to an earlier version Keep my code completely private by running everything on my own computer Use my own AI model account instead of paying a subscription

README

Empryo

Empryo

previously SoulForge

The AI coding agent that edits symbols, not strings.

Empryo reads your repository as a living system: it sequences a genome — a live, ranked dependency graph of every symbol — and edits through the AST, never find-and-replace.

Website · Download · Benchmarks · Changelog · Discussions · Discord

Empryo in action

SoulForge is now Empryo — the same symbol-level, graph-powered agent, rebuilt with a desktop app, a faster engine, and a composable core. This repository is Empryo's public home for issues and discussions.

Install

# macOS / Linux
curl -fsSL https://empryo.com/install.sh | bash

# Windows (PowerShell)
irm https://empryo.com/install.ps1 | iex
brew install proxysoul/tap/empryo
winget install ProxySoul.Empryo
empryo --set-key anthropic sk-ant-...   # or run locally with Ollama — no key required
cd your-project
empryo

Desktop app and prebuilt binaries: empryo.com/download. Runs on macOS, Linux, and Windows.

Why Empryo

Most coding agents grep, read whole files, and patch strings — they never know what depends on the code they just changed. Empryo builds understanding before it mutates anything:

  • It maps before it reads. On launch, tree-sitter parses your repo into a live graph — every symbol, import, and call site, ranked by PageRank and git co-change. Graph queries answer in milliseconds and cost zero LLM tokens.
  • It knows the blast radius. Before an edit, the agent sees what imports a file, what historically changes with it, and how far a change ripples — "what breaks if I touch this?" is answered before the first keystroke.
  • It edits through the AST. 65+ symbol-level operations, atomic batches with all-or-nothing rollback, structural edits across 30+ languages, and a typecheck as the gate. Nothing breaks on whitespace.
  • It treats tokens as spend. The graph does the navigation models usually burn context on — fewer reads, fewer steps, smaller bills at any scale.

What's inside

Empryo FAQ

Which apps does this work with?
Empryo is its own standalone application rather than a plugin for another editor — it comes as a desktop app, a full in-terminal interface, and a plain command-line mode you can run from scripts.
Do I need an account key to use it?
Usually yes — you connect your own account from a provider like Anthropic (Claude), OpenAI, or Google and pay only for what you use. If you'd rather not use an outside provider at all, you can run it completely free with a local, offline model instead.
How hard is it to set up?
You install it with one line in your terminal, or with a package manager on Mac or Windows, then run one command to add your account key. It's a real installation rather than a single click, so it suits someone comfortable typing a couple of commands.
Can I use this to fix bugs in my own project?
Yes — it reads your whole project first, so when it fixes a bug it also checks what else depends on that code before changing anything.
Can I use this to understand a messy codebase I didn't write?
Yes — it builds a map of how every file and function connects, so you can see what matters and what touches what before you change anything.
Does it cost money to use?
The application itself is free. You only pay for use of whichever AI provider account you connect, such as Anthropic, OpenAI, or Google, or you can skip paying entirely by running a local model.
Can it undo changes it makes?
Yes — every step is saved like a checkpoint, so you can rewind your code and the conversation back to any earlier point.
Does my code leave my computer?
No — it runs entirely on your own machine, with no separate company server sitting in between you and your code.