Before
Vague follow-ups, copied corrections, and lost context between agent turns.
Native macOS workspace for CLI agents
Rally runs Codex, Claude Code, and Gemini CLI in one native workspace, so you can review agent output, mark exactly what needs to change, and send it back as concrete instructions.
Works with
Why Rally
Vague follow-ups, copied corrections, and lost context between agent turns.
Select the exact span, attach notes, collect them, and send one concrete bundle.
Input Harness
Rally ties feedback to the agent output it came from. The next turn gets specific context instead of another vague prompt.
Mark the exact span of agent output that needs attention.
Attach a focused note to that selected range.
Gather multiple notes across one response.
Deliver them as a single instruction bundle.
Run Codex, Claude Code, and Gemini CLI side by side.
Keep sessions, previews, and local documents organized.
Use a real shell powered by xterm.js and node-pty.
Open local previews and links beside the agent work.
Manage local MCP servers and skill folders from settings.
No hosted Rally service, no prompt proxy, no model access.
How it works
Local-first and transparent
Rally runs on your machine and launches the command line agents installed there. Coding agents can run commands and edit files based on their configured permission mode, so use trusted repositories and review permissions carefully.
Download
Apple Silicon · macOS 12.0+ · v0.0.1 · Free beta
Thanks — your download is starting.
Download againSHA-256: 21e07fd1ec87be486162bae8b43cdedbde0d94a62b359ffe91e219d2915f52a9
Open the DMG, drag Rally to Applications, then right-click Rally and choose Open.
Signed and notarized builds, Homebrew Cask, and auto-update are planned next.