Logo
Deploy Now

Categories

Self-Hosted Developer Tools DevOps AI Automation Security Code Review

Stars

259

Forks

45

Watchers

2

Developer links

Last Updated: 08-24-2026 11:51 PM (UTC)

Mira

The fastest tool on the public Martian Code Review Bench at 77 seconds per PR with competitive F1 scores — while every higher-scoring competitor takes 5 to 14 times longer — Mira is the fully open-source AI code reviewer that ships the entire feature surface other tools gate behind per-seat subscriptions: indexed PR reviews with full-repo context, streaming walkthroughs posted within seconds, inline vulnerability scanning from hourly OSV.dev polls, org-wide package inventory answering "which repos use [email protected]?" in one query, interactive dependency graphs with blast-radius SVG rendering via ReactFlow, a learning loop that synthesizes rules from rejected comments and human review patterns on merged PRs, and cost telemetry showing actual spend per repo and per model. Unlike CodeRabbit, Greptile, and GitHub Copilot code review which are SaaS-only with per-seat pricing, Mira runs as a single Docker image on your infrastructure with your LLM key through OpenRouter (fronting Anthropic, OpenAI, Google Gemini, DeepSeek) or direct to Ollama and vLLM for fully air-gapped deployments. The engine indexes Python, TypeScript, Go, Rust, Java, C/C++, C#, Swift, Kotlin, Scala, and PHP with per-file symbol extraction, enabling multi-file reasoning across diffs rather than isolated line-by-line review. Integrates with GitHub, GitLab, and Forgejo via webhook with full feature parity. Running on a dedicated VPS on RepoCloud with guaranteed CPU, RAM, and SSD, full root SSH access, and a browser serial console. Apache 2.0 licensed.

Mira
Mira
Mira

Benefits

  • Fastest Benchmarked Code Reviewer
  • Completes full PR reviews in approximately 77 seconds while maintaining competitive F1 accuracy scores on the public Martian Code Review Bench against per-seat SaaS competitors.
  • Full-Repository Context Indexing
  • Indexes entire codebases with per-file symbol extraction and dependency graphs, enabling cross-file reasoning about call graphs, data flow, and blast radius rather than diff-only analysis.
  • Zero Vendor Lock-In Architecture
  • Runs any LLM via OpenRouter, direct provider APIs, or local models through Ollama and vLLM, with separate model configuration for cheap indexing versus powerful review passes.
  • Self-Improving Review Quality
  • Learning loop synthesizes custom rules from rejected comments and human review patterns on merged PRs, continuously reducing noise and improving relevance over time.

Features

  • Streaming PR Walkthroughs
  • Posts placeholder comments within one second, narrative summaries within ten seconds, and complete inline reviews with severity scores within a minute.
  • Vulnerability Scanning
  • Hourly OSV.dev polling surfaces CVE alerts across every package in every indexed repository with severity, advisory links, and fix versions inline.
  • Multi-Platform Integration
  • Full feature parity across GitHub, GitLab, and Forgejo including inline comments, bot mentions, and webhook-driven automatic review triggering.
  • Analytics Dashboard
  • React dashboard showing per-repo review stats, token spend, issue severity breakdowns, reviewer responsiveness leaderboards, and rubber-stamp detection.
  • Custom Rules Engine
  • Define org-wide or per-repo review rules via the dashboard or .mira.yaml configuration files, enforced alongside LLM-generated findings.