Repository architecture mapping
Hyrax maps modules, entry points, ownership, dependencies, and architectural boundaries so structural problems can be examined in repository context.
Hyrax
Hyrax analyzes software repositories, identifies architectural and code-health issues, reviews pull requests, and generates verified fixes for human approval. Its MCP server gives Claude Code, Cursor, and Copilot live access to repository findings and rules while developers work in their editors.
Codebase analysis, fixes, and review
Hyrax is a codebase analysis and improvement tool for AI-native engineering teams. It maps repository architecture and dependencies, audits code for issues, reviews pull requests, and turns selected findings into verified changes delivered as pull requests.
The product is designed to keep automated coding work aligned with repository rules while leaving approval and merging with engineers. Its MCP server extends this workflow into Claude Code, Cursor, and Copilot by providing live findings, conventions, and job status instead of relying only on static context files.
Core capabilities
Codebase audits
Audits identify issues across security, correctness, maintainability, performance, architecture, and operations, rather than limiting review to AI-generated code.
Verified fixes
For supported findings, Hyrax produces verified changes as merge-ready pull requests. The documented workflow includes test and production-build checks before review.
Pull-request reviews
Hyrax reviews pull requests and posts review comments in GitHub, allowing teams to see findings in their existing code-review workflow.
Live editor context
The MCP server exposes live findings, applicable rules, canonical patterns, pending migrations, cost forecasts, and job status to Claude Code, Cursor, and Copilot.
Governed automation
Teams can submit audits and fix jobs, inspect results, and triage findings by dismissing, acknowledging, completing, or reopening them. API-key scopes and spend controls govern these actions.
Practical use cases
Understand an unfamiliar codebase
Before changing a mature repository, a team can inspect modules, entry points, ownership, dependencies, and layering violations to identify the cause of an architectural problem rather than only its symptom.
Prioritize code health work
An engineering team can run an audit across security, correctness, maintainability, performance, architecture, and operations, then select findings for remediation based on the repository's current state.
Turn findings into controlled fixes
When a finding has a clear remedy, Hyrax can implement the change, run documented validation such as tests or a production build, and open a pull request for an engineer to review.
Review changes in the existing workflow
Teams can use Hyrax's GitHub pull-request reviews to add independent feedback to normal development, including code that was not written by an AI system.
Give coding agents live repository context
A developer working in Claude Code, Cursor, or Copilot can ask which findings and conventions apply to a file, submit a related job, and track the resulting pull request without leaving the editor.
Pros and Cons
Pros
- Combines architecture mapping, audits, pull-request reviews, and verified fixes in one workflow.
- Uses pull requests as the controlled output for changes, preserving human review and merge approval.
- Provides live repository findings and rules to Claude Code, Cursor, and Copilot through MCP.
- Free workspaces receive the full product rather than a feature-limited tier, with usage controlled by credits.
- Supports scoped API keys, IP restrictions, spend controls, and optional paid-plan overage caps.
Cons
- The available evidence does not document the full repository-language matrix or the complete range of supported audit checks.
- MCP access requires API access, and audits, fixes, and reviews can consume metered credits; work stops when credits run out unless a paid workspace enables capped overage.
- Fixes remain subject to human review and merge, so Hyrax does not provide an auto-merge workflow.
FAQ
What does Hyrax do?
Hyrax maps repository modules, entry points, ownership, dependencies, and architectural problems, then can propose or implement a targeted change. Its documented workflows include repository audits, verified fixes, and pull-request reviews.
Does Hyrax automatically merge fixes?
Fixes are delivered as pull requests. A human engineer reviews and merges the result; Hyrax does not auto-merge changes.
Which coding tools can use Hyrax's MCP server?
Hyrax provides an MCP server over streamable HTTP. Claude Code, Cursor, and Copilot can use it to query live findings and rules, submit audit or fix jobs, and check job or pull-request status. MCP access requires API access and uses the same API-key authentication model as the REST API.
How does Hyrax pricing work?
Free includes the full product, a $30 starter credit, $10 per month in ongoing credits, and up to 100 PR reviews per month at no charge. Audits, fixes, and reviews draw on credits when applicable. Paid costs $30 per user per month and provides $30 per month of credits per paid user.
How are code privacy and usage limits handled?
All inference runs in Hyrax's AWS Bedrock account, and the site states that customer code is not used for model training. Paid workspaces can also enable optional overage and set a budget cap.
Quick Facts
- Category
- Developer tool
- Primary users
- AI-native engineering teams
- Core workflows
- Repository audits, verified fixes, and pull-request reviews
- Editor integrations
- Claude Code, Cursor, and Copilot via MCP
- Change delivery
- Merge-ready pull requests with human approval
- Pricing model
- Free plan plus $30 per user per month paid plan; usage uses credits
Hyrax Alternativen
ByteAsk
ByteAsk is a terminal-first AI coding agent for C and C++ that edits repositories and verifies changes with the real compiler, debugger, sanitizers, and tests before showing a diff. It offers a free tier plus paid plans, with editor connectors and zero-retention handling described in the source.
GitBot
GitBot is a local browser-based workspace for creating reusable coding bots powered by Claude Code, Codex, or OpenCode. It helps developers run those agents across repository folders with persistent threads and configurable permissions.
Manta AI
Manta AI is an autonomous web app testing tool for teams that want to map application behavior, catch regressions, and generate tests without writing scripts or maintaining selectors. It works from a URL and supports plain-English test flows, run results with screenshots, and scheduled or deployment-triggered checks.
Ghost
Ghost ist ein terminalbasierter KI-Assistent für Chats, Code-Generierung und Aufgaben direkt in der Kommandozeile. Mit kostenlosen Modellen, Linux, macOS und Windows. Open Source.
Vi3ecode
Vi3ecode is a maintained development environment for working with AI coding agents. It keeps project context, terminal, Git, memory, flows, review, and communication connected around the agent you already use.
Lucid Train
Lucid Train is a local-first AI coding harness that turns repository code into architecture diagrams and uses those diagrams as specifications for coding agents. It is available as a desktop app and a lightweight Rust CLI, and it can run with local models or your own API keys.