plannotator-last

Annotate the latest assistant message in Plannotator and apply returned feedback.

9|Updated Jul 17, 2026
One-click install
npx skills add https://github.com/Yassimba/loom --skill plannotator-last-yassimba
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: plannotator-last
Source: https://github.com/Yassimba/loom/tree/main/drafts/plannotator-last
Command: npx skills add https://github.com/Yassimba/loom --skill plannotator-last-yassimba

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires plannotator.

What problem does it solve? When reviewing an AI assistant's most recent response, users need a way to visually annotate that exact message and feed the annotations back into the conversation without manually copying text or re-explaining context. ## Core Features & Use Cases - Targeted Annotation: Runs plannotator last to open Plannotator on the latest rendered assistant message, avoiding accidental annotation of preamble or status messages. - Feedback Incorporation: Reads the returned annotation session result and revises the message or continues work based on the feedback. - Approval Notes Handling: Recognizes that an approved decision may still carry guidance notes, carrying them into subsequent work without redoing the message. - Use Case: After the assistant drafts a plan, invoke this skill to mark up that plan in Plannotator, then have the assistant revise the plan according to your annotations. ## Quick Start Ask the assistant to open the latest response in Plannotator so you can annotate it and have the feedback applied.

Frequently Asked Questions about plannotator-last

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I annotate the latest assistant message with Plannotator?

Invoke this skill, which runs the plannotator last command to open the most recent rendered assistant response in Plannotator. After the annotation session finishes, the assistant incorporates any returned feedback into its follow-up.

What happens if I approve a message but add notes in Plannotator?

An approved decision can still include a feedback field with notes. The assistant reads those notes and carries them into subsequent work as guidance, without redoing the approved message.

Why should I avoid sending a message before running plannotator last?

The command targets the latest rendered assistant response, so a commentary or status message sent beforehand could become the thing being annotated instead of the intended response.

What happens if the Plannotator session closes without feedback?

If the annotation session ends without returning feedback, the assistant briefly mentions that and continues with the work rather than blocking or retrying.

What tool does this skill depend on?

It depends on the plannotator CLI tool, declared in its deps.yml file. Installers pull this dependency in transitively so the plannotator last command is available to run via Bash.