Keyboard-Lord

Keyboard-Lord / Rasputin-Coder

Public

Rasputin is a terminal-native, local autonomous software engineering agent powered by Qwen-Coder. It plans, executes, validates, and self-heals code changes with a deterministic audit trail, checkpoint recovery, and zero cloud dependency — a true “Codex at home.”

10
1
100% credibility
Found Apr 23, 2026 at 10 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

Rasputin is a terminal coding companion that uses local AI to plan and safely execute code edits with validation gates.

How It Works

1
🖥️ Launch in your code folder

Open your terminal, go to your project, and start the coding helper with one simple command.

2
🤖 Connect local thinker

It automatically finds and links to your nearby AI brain running on your computer.

3
💭 Tell it what to fix or add

Type a plain request like 'fix the math bug' or 'add a new button' just like chatting.

4
📋 See the safe plan

Review the step-by-step changes it proposes, with risks and files highlighted.

5
🔧 Approve and let it work

Give the okay, and it edits files, runs checks, and tests everything safely.

✅ Code improved, ready to use

Your changes are done, tests pass, and you see diffs and logs of what happened.

Sign up to see the full architecture

4 more

Sign Up Free

Star Growth

See how this repo grew from 10 to 10 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is Rasputin-Coder?

Rasputin-Coder is a terminal-native autonomous software engineering agent powered by Qwen-Coder, running locally via Ollama with zero cloud dependency. Developers input natural-language goals like "fix parser routing," and it plans steps, executes code changes, validates builds/tests, and self-heals failures through bounded chains. Rust-built, it provides deterministic audit trails, checkpoint recovery, and replay for tracing every autonomous action.

Why is it gaining traction?

Its fully local execution avoids cloud costs and data leaks, while the Forge runtime bounds agent behavior to prevent infinite loops—validating lint/build/test stages before committing changes. The audit log and replay features let you debug agent decisions deterministically, unlike black-box cloud agents. Early benchmarks on bug fixes and refactors show reliable self-healing without external dependencies.

Who should use this?

Rust devs prototyping CLI tools or backend services solo, needing quick autonomous code edits without cloud APIs. Indie hackers iterating on open-source repos who want audit trails for every agent-executed change. Terminal power users tired of VS Code extensions for local LLM coding.

Verdict

Promising local agent for bounded autonomous coding (10 stars, 1.0% credibility score), with excellent docs and benchmark workspaces—but too early for production; test on toy repos first. Solid foundation if you need cloud-free, auditable code execution.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.