Plannotator

Plan and code review workspace for AI coding agents

Visit website →

About

Plannotator is an open-source, browser-based review workspace for AI coding agents including Claude Code, Codex, Copilot, Gemini CLI, and OpenCode. Instead of squinting at agent plans in the terminal, you get a proper review surface where you can annotate, comment, and own the specification.

When your agent proposes a plan, Plannotator intercepts the approval step and opens a review UI. You select text, mark deletions, add comments, or write replacements. Your annotations export as structured feedback the agent understands and acts on.

It also supports PR-style code review with side-by-side diffs, file tree navigation, and line-level annotations. Everything runs locally — your review data stays on your machine by default.

AI-assisted draft, human-reviewed before publishing — see how we choose & review tools.

Why we picked it

It solves a real pain point in agent workflows: how do you review what an agent plans to do before it does it? Instead of terminal text, you get a proper review surface. Open-source and already at 12k stars.