Plannotator
Annotate, review, and share Claude Code plans. (Plugin/Hook)
128 followers
Annotate, review, and share Claude Code plans. (Plugin/Hook)
128 followers
Interactive Plan Review: Mark up and refine your plans using a UI, easily share for team collaboration, automatically integrates with agent plan mode. Select the exact parts of the plan you want to change. Mark it for deletion, add a comment, or suggest a replacement. Share plans and collect team member feedback. Automatically send feedback for Claude Code to act on. Runs locally. Local plugin. No network requests. Plannotator runs entirely in your browser. Plans never leave your machine.
This is the 2nd launch from Plannotator. View more
Plannotator
Launching today
You can now use Plannotator to annotate anything! Free, open source, and private (runs local).
Use the skill `/plannotator-annotate` (or cli) - followed by a folder, URL, document, markdown, your book, a recipe, a plan, etc - to annotate anything and send feedback directly into your integrated agent.
Use `/plannotator-last` to annotate the last message an agent sent you too.
I like using it for my own version of Karpathy’s "LLM Wiki" in obsidian.
https://github.com/backnotprop/plannotator
Free
Launch Team / Built With
Plannotator
I wanted to be able to annotate anything and have feedback integrated directly into the agent loop. That's why I created this. It was also a constant desire to annotate the last agent message. Happy to introduce both capabilities in this new product launch.
Free, open source, your work runs locally and stays private.
https://github.com/backnotprop/plannotator
Try a demo of the annotation surface here: https://share.plannotator.ai
OpenPlugin
the plan-annotation-as-feedback-loop is the part that's been missing from claude code workflows. most people are still pasting raw plan text into a doc, marking it up by hand, then re-pasting back into the agent, which is brutal and the round-trip kills the energy. the question for me is whether the team-feedback-collection becomes async (collect + bulk-apply) or stays synchronous like a code review thread. the async version unlocks distributed teams but the synchronous one is where the actual disagreements get resolved.
This is a really clever take on closing the loop between humans and agents .Annotating directly feeding that back into the agent feels like a missing piece.