Mira - Code Review

Self-hosted AI code review. Your code, your dashboard, your LLM key.

Self-host every feature: full review engine, codebase indexing, vulnerability scanning, custom rules, org-wide package search, dashboard, learning loop. No paid tier, no license key, no SaaS upsell.

Mira reviews your pull requests using your choice of LLM (via OpenRouter, which fronts Anthropic, OpenAI, Google, DeepSeek, and more) and posts concise, actionable feedback. The noise filter, confidence clamping, and learning loop ensure you only see comments that matter.

Highlights:

• Any OpenAI-compatible endpoint: OpenRouter by default (Anthropic, OpenAI, Google Gemini, DeepSeek, …), or point llm.base_url at vLLM, Ollama, LiteLLM proxy, LocalAI, llama.cpp, Together, Fireworks, Groq, or Cerebras for self-hosted / data-resident reviews. See Custom endpoints.
• Low noise: Confidence thresholds, dedupe, severity sorting, per-PR comment caps
• Indexed reviews: full-repo code index gives the LLM real context, not just the diff
• Learns your team: synthesizes rules from rejected comments and human review patterns on merged PRs
• Vulnerability scanning: hourly OSV.dev poll surfaces CVEs across every package in every repo
• Org-wide package search: answer "which repos use lodash@4.17.20?" in seconds
• Configurable: .mira.yaml for per-repo settings, custom + global rules in the dashboard
• Self-host on day one: Docker image, Railway / Fly.io / Render configs, SQLite or Postgres
• GitHub today, more platforms coming: GitLab, Bitbucket, and Gitea support are on the roadmap. Same review engine, same dashboard, just a different provider adapter