Get started in minutes.
Review pull requests through the GitHub App, run Mindrealm locally from your terminal, and your agent's stop-hook.
Review pull requests
Install the GitHub App
Add Mindrealm to your organization and pick the repositories it should review. No CI configuration to write.
Open a pull request
Mindrealm clones the branch, reviews every change, and posts a check run with findings annotated on the exact lines that need them.
Fix findings and push again
Using the mind review CLI or the Claude Code, Codex, Kimi Code, or Antigravity hook, automatically address the findings and push again.
Approved teams onboard directly with the founder.
Run it locally
Install the mind CLI on macOS or Linux:
curl -fsSL https://mindrealm.ai/install.sh | shOn Windows, in PowerShell:
iwr -useb https://mindrealm.ai/install.ps1 | iexOr download a binary for your platform from the releases page, then extract it and put mind on your PATH.
Re-run the same command any time to update.
Then authenticate:
mind loginFrom inside a repository connected to your GitHub App, review your changes:
mind reviewWire it into Claude Code, Codex, Kimi Code, or Antigravity so it reviews automatically whenever your agent finishes. One command detects your agent and sets up the auto-fix loop:
mind setupDisagree with a finding? Silence one check with a mind:disable comment on the line above it. It stays recorded and reviewable, never hidden:
// mind:disable (Go, TypeScript, Rust)
# mind:disable (Python)
// mind:disable unused-export (silence just one rule)See the docs for the full configuration reference, every command, and suppressions.