penberg

penberg / rp

Public

Repair programs

ai-coding-agent bug
19
0
100% credibility
Found Mar 23, 2026 at 19 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Rust
AI Summary

rp is a tool that helps developers automatically analyze bug reports, create local tests to reproduce issues, and apply AI-powered fixes to software projects.

How It Works

1
🔍 Spot a bug

While working on your software project, you find a bug report describing a problem others are facing.

2
📥 Bring in the bug fixer

You add the rp tool to your project folder to help automatically repair issues.

3
🎯 Show the bug report

You share the details of the bug report with rp so it can study the problem.

4
🧪 rp creates a bug test

rp thinks carefully and makes a simple test that repeats the bug exactly on your computer.

5
🔍 Run the test

You try the test to confirm it shows the bug failing just like expected.

6
🛠️ rp repairs the code

You ask rp to fix it, and it smartly updates your code to solve the issue.

🎉 Everything works!

The test now passes perfectly, the bug is gone, and your project is healthier than ever.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 19 to 19 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 rp?

rp is a Rust CLI tool that repairs programs by structuring AI coding agents like Claude, Codex, or OpenCode to tackle bugs. Feed it a GitHub issue URL via `rp inspect`, and it generates a local POSIX shell reproducer script, a summary, and analysis notes—turning vague reports into deterministic failures you can `rp check` offline. Then `rp fix` invokes the agent to patch code until the repro passes plus your project's verify command, like `cargo test` or `make test`.

Why is it gaining traction?

It bridges AI agents' power with reproducible local workflows, avoiding flaky cloud replays—inspect once, fix repeatably without GitHub pings. Auto-detects project types for init configs, enforces regression tests in designated roots, and handles car repair programs or image repair github scenarios by focusing on verifiable fixes. Devs dig the agent-agnostic setup and bash repros that exit non-zero on bugs.

Who should use this?

Open-source maintainers fixing GitHub issues in Rust, Node, Python, or Go repos, especially for github repair pdf, github repair video, or jpeg repair github edge cases. Solo devs wanting direct repair programs for github boot repair, mac repair programs, or repair windows update bugs without manual repro hunting. Teams integrating AI for repair office programs or repair programs apple in monorepos.

Verdict

Worth a spin for AI-curious maintainers—solid README, installer script, and blog post make onboarding fast, but 19 stars and 1.0% credibility score signal early maturity with room for broader agent support and battle tests. Pair it with your workflow if you're tired of issue triage drudgery.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.