[FEATURE] In-line Comments on Claude Artifacts

Preflight Checklist

• [x] I have searched existing requests and this feature hasn't been requested yet
• [x] This is a single feature request (not multiple features)

Problem Statement

As we move up the value chain we’re basically exclusively interacting over Claude Artifacts (not looking at code, not looking at PRs). But we can’t leave in-line comments on artifacts (so that claude/other engineers can see them and iterate on them). Being able to leave comments and feedback on artifacts would really speed everything up and level up scale.

This has been requested in an invalid/malformatted way here:
https://github.com/anthropics/claude-code/issues/86479

and again (similar, maybe duplicate depending on interpretation) here:
https://github.com/anthropics/claude-code/issues/68019

Proposed Solution

In-line comments similar to google docs that Claude and other users can interact with. Claude should be able to reformat the artifact based on feedback similar to updating a PR based on feedback.

Alternative Solutions

Currently we work around this by taking these discussions and feedback into Slack threads or PR comments and then pass those back to Claude to have it rework the artifact

Priority

High - Significant impact on productivity

Feature Category

Other

Use Case Example

Example Scenario:

1. I'm working on a massive redesign of our financials architecture.
2. Fable builds an artifact outlining the architecture, and an implementation plan of how the work will be distributed among agents
3. Experts on our team in insurance financials leave feedback on the architecture, they have perspectives on our customers and the business side that Claude lacks.
4. Experts on our engineering team leave feedback on the architecture, they have eng perspectives that it's important for claude to take into account
5. Where the two overlap, they discuss and resolve in comments
6. Claude continuously updates the artifact while discussion is had an comments are left, only updating the parts of the doc that need to be updated based on feedback
7. implementation begins and Claude kicks off suites of agents to implement the new feature and architecture

Additional Context

This has been asked for in many other forms