Niches · Design workspaces & canvases
Act on review comments left on a design
One company's docs show a full fit.
What counts as done
Each open thread on the named file read with the node it is pinned to, the change made, and a reply posted on that thread, with comment text treated as data.
How agents ask for it: address the comments on this file
, reply to the design feedback
, what did the reviewer ask for on this frame
Does the job 1
The docs show it meets the test above, and nothing we read says otherwise.
Does the job, with a limit 7
It does the job, but something in the docs or a user's account will get in an agent's way. The limit is named in each row.
| Listing | Why |
|---|---|
| Figma API + MCP B Figma, Inc. | MCP has no comment tools. The REST API lists comments, posts replies to root comments with client_meta positions, and deletes, but has no resolve call, so the agent needs a second credential. source |
| Melius C Melius AI, Inc. | comment_list, comment_reply and comment_resolve exist, but threads are created at a canvas position and the docs don't say a listed comment names the node it sits on. source |
| Miro API + MCP B RealtimeBoard Inc. dba Miro | comment_list_comments reads comments for a board or item, with reply and resolve, and canvas_update_from_svg makes the change. Board comments are a whiteboard's, not a design file's, and no prompt-injection guidance is given. source |
| Penpot API + MCP F Kaleidos Subsidiary SL | The plugin API behind execute_code has findCommentThreads, the board a thread is attached to, reply and a resolved flag, so threads can be handled. Whether the MCP plugin holds comment permissions is not stated. source |
| Subframe E Atomic Design Inc | list_comments returns open threads with the element id they sit on, so edit_page can act on them, but replying and resolving happen only in the Subframe app. source |
| UX Pilot MCP server F UX Pilot | read returns comments and manage replies and resolves them, with version_from_comments to apply feedback. Whether a comment carries the element it is pinned to isn't stated. source |
| Zeplin D Zeplin, Inc. | Screen notes with positions and their comments can be listed, created and updated through REST, but notes are pinned to screen coordinates, not layers, and the design itself can't be changed in Zeplin. source |
Couldn't check 1
The docs were closed to us, or silent. These carry no mark either way.
| Listing | Why |
|---|---|
| Sketch D Sketch B.V. | Sketch has document comments, but no MCP tool for them is listed and the docs don't say whether SketchAPI through run_code reaches them. |
Doesn't do this job 8
Listings in the category that do something else, with the reason.
| Listing | Why |
|---|---|
| ArtCraft F Storyteller | No comments. source |
| Framelink MCP for Figma B Framelink (Graham Lipsman) | No comment tool; only design data and images. source |
| Framer Server API C Framer B.V. | No comment reading or replying is described for MCP or the Server API. source |
| Impeccable BB Renaissance Geek, Inc. | No design-file comments. source |
| Magic Patterns B North Park Labs, Inc. | No comment tools in the MCP tool list. source |
| pen.dev E High Agency, Inc. | MCP exposes no comments; acc-23 is open asking for it. source |
| Presentations.AI MCP server E Presentations.AI | No comment tools. source |
| UXKIN C UXKIN (no legal entity named) | No comment tools. source |
What stops agents finishing
- comments stored server-side and not exposed to the agent
Around it
Before: design.edit-canvas. After: design.edit-canvas.
Where this comes from
Weight 4 of 100 in Design workspaces & canvases: 1 of 42 accounts (pen.dev issue) plus vendor pages for Kite @-mentions and Figma get_comments; weight is a guess
The same as data: /api/v1/use-cases/design.review-comments.json, or the MCP tool get_use_case with id design.review-comments. Other jobs in this category: /api/v1/niches?category=design.
For companies
Do agents find, use and choose your tools?
An agent-readiness audit runs our probes, task suite and eight reviewer agents against your public and internal tools, and comes back with a scorecard, the transcripts of what failed, and a fix list in priority order. From $2,500, re-run included. We never take payment to move a rank. We do help companies earn one.