Refraction
refraction.devRefraction uses AI to refactor code, generate documentation, and create unit tests from code samples.
Looking for something like OpenCodeReview? Here are 17 alternatives, ranked by how closely they match it on purpose, features and platforms. Of these, 11 are open source.
OpenCodeReview combines deterministic checks and language model agents for code review with line-level comments.
| # | Alternative | Pricing | Open source | Platforms |
|---|---|---|---|---|
| 1 | Refraction | — | Not stated | Web |
| 2 | LangWatch | — | Yes | — |
| 3 | PR-Agent | — | Yes | — |
| 4 | ReviewCerberus | — | Yes | — |
| 5 | PR-Agent | — | Yes | — |
| 6 | Blinky | — | Yes | — |
| 7 | rehydra | — | Yes | — |
| 8 | EvalView | — | Yes | — |
| 9 | Greywall | — | Yes | Linux, macOS |
| 10 | promptfoo | — | Yes | — |
Refraction uses AI to refactor code, generate documentation, and create unit tests from code samples.
LangWatch provides evaluations for language models and tests for AI agent applications.
This repository contains code for PR-Agent, an open source tool for reviewing pull requests.
ReviewCerberus analyzes Git branch differences and generates code review reports using AWS Bedrock or Anthropic API.
PR-Agent provides automated review of software pull requests in this open source repository.
Blinky provides an open-source AI debugging agent integrated into Visual Studio Code.
rehydra scans LLM prompts for accidental personal information before they reach a model.
EvalView regression tests AI agents by snapshotting behavior and comparing tool calls in CI.
Greywall provides a deny-by-default sandbox for AI coding agents, with kernel-enforced isolation on Linux and macOS.
promptfoo tests prompts, agents, and retrieval systems with evaluations and security testing.
h5i provides a red-teaming workspace with browser access, HTTP traffic controls, sandboxed execution, and auditable sessions.
LangKit monitors language model prompts and responses and extracts signals related to safety and security.
Kodus provides open-source AI code review that teams can self-host or run in the cloud using their own model keys.
Prediction Guard provides a self-hosted platform for applying NIST and OWASP policies to models and agents.
Snyk DeepCode AI helps review code and identify security issues, with autofix features for developers.
Qodo reviews code with codebase context and supports workflows for IDEs, pull requests, and security.
Kiro turns prompts into software requirements, architecture, and task plans, then uses agents to implement and test code.
Alternatives are suggested from the purpose and features described in our listings. Different pages for the same product appear once. Read each description to check whether it fits your task. How we choose alternatives · Suggest an alternative