CLAUDE_CODE_ARTIFACT_COMMENTS
A Claude Code environment variable, read out of the shipped bundle. It has been in the build since v2.1.221, including the newest one read.
What it is
Set to 0 to stop Claude reading and replying to [comments on an artifact](/docs/en/artifacts#collect-comments-on-an-artifact). Has no effect when CLAUDE_CODE_DISABLE_NONESSENTIAL_TRAFFIC has [turned artifacts off](/docs/en/artifacts#availability). Requires Claude Code v2.1.221 or later
Anthropic's own wording. Read off Environment variables, captured 2026-08-28.
Presence across releases
| Releases | Reading | Declared type | Default |
|---|---|---|---|
| 2.1.245 – 2.1.250 5 releases | never mined | — | — |
| 2.1.221 – 2.1.243 22 releases | in the build | triBool | — |
A row reading never mined is a release with no archived bundle behind it. Nothing was read out of it, so nothing here says whether the name was in it. A declared type or a default is only ever what that release's own bundle stated.
Changelog entries naming it
-
v2.1.248Artifact comment threads are summarised before Claude acts on them
Artifact comment threads get summarised so the permission prompt names what's being acted on.
named in this entry
-
v2.1.248Permission prompts say whose comment is being acted on
Comment permission prompts can name whose comment it is, behind an artifact-comments env var.
named in this entry
-
v2.1.246Artifact comments refuse organization service keys
Artifact comments now explain org service key failures clearly, but comments themselves need an env var.
named in this entry
-
v2.1.239Artifact request schema: auto_reply gated, embeds compiled out
Artifact auto-reply appears only with comments enabled, and an embeds field is compiled out entirely.
named in this entry
-
v2.1.239Artifact watch entries gain unread-comment counters behind the comments flag
Artifact watches can track unread comments and replies, but only with the comments flag on.
named in this entry
-
v2.1.235Artifact MCP is on by default
The built-in artifact MCP server is now on for everyone unless you set the env var to disable it.
named in this entry
-
v2.1.227Artifact auto-replies stop describing their own plumbing
Automatic artifact comment replies now sound like plain acknowledgements instead of describing internal plumbing.
named in this entry
-
v2.1.227Artifact tool can resolve comment threads, behind an off-by-default flag
Claude could mark a comment thread on a published artifact as resolved, but the path is off.
named in this entry
-
v2.1.221Artifact comment errors distinguish egress-proxy blocks and non-activated threads
Artifact comment failures would name the real cause, like a proxy block or an unactivated thread, once enabled.
named in this entry
-
v2.1.221Artifact live-update arming carries auto-react wiring
Artifact live-update watching now carries the plumbing needed for automatic comment replies and cancellation.
named in this entry
-
v2.1.221Comment replies always ask, and the prompt names who will see them
Replying to an artifact comment always asks first, and tells you whether the artifact is shared.
named in this entry
-
v2.1.221Artifact auto-reply subscriptions can be stopped from the cancel gestures
Escape or double ctrl+x ctrl+k now also turns off automatic replies to comments on your artifacts.
named in this entry
-
v2.1.221New artifact comment threads: read, reply, resolve
Reading, replying to and resolving comment threads on published artifacts is built in but off unless enabled.
named in this entry
Documentation pages
- Environment variables reference table Claude Code CLI
A reference table page is one whose own table defines this name, and it is where the description above came from. A page that mentions it carries the name somewhere in its text and may say nothing about it at all. Every page carrying it.
Names beside it
Every environment variable in the inventory starting claude_code_artifact, up to twelve of them.
CLAUDE_CODE_ARTIFACTPR review artifact pages can show a pinned button that approves the pull request from your GitHub account. 2.1.166
CLAUDE_CODE_ARTIFACT_ASSET_BASE_URLThree artifact environment variables are now recognised, with the API token stripped from spawned processes as a secret. 2.1.221
CLAUDE_CODE_ARTIFACT_ASSETSClaude can now upload, list, read and delete images, video, PDFs and fonts attached to your published artifacts. 2.1.234
CLAUDE_CODE_ARTIFACT_AUTO_OPENSet to 0 to stop Claude Code from opening the browser automatically when a new [artifact](/docs/en/artifacts) is published. Republishing an existing artifact does not open the browser regardless of this setting 2.1.172
CLAUDE_CODE_ARTIFACT_COMMENT_FAST_ACKClaude can post a quick 'on it' note in artifact comment threads before the real reply, off unless enabled. 2.1.232
CLAUDE_CODE_ARTIFACT_COMMENT_FAST_ACK_FIXEDA cheaper canned way to acknowledge artifact comments exists but needs two env vars switched on. 2.1.235
CLAUDE_CODE_ARTIFACT_COMMENT_RESPONDERA read-only subagent that analyses one artifact comment thread exists but stays off. 2.1.227
CLAUDE_CODE_ARTIFACT_COMMENTS_AUTOREACTSet to 0 to stop Claude [replying on its own to comments sent to it](/docs/en/artifacts#let-claude-reply-to-comments-on-its-own). Requires Claude Code v2.1.228 or later 2.1.221
CLAUDE_CODE_ARTIFACT_DBPublished artifacts can gain a document database Claude reads and writes, off unless you set the env var. 2.1.224
CLAUDE_CODE_ARTIFACT_DELETEClaude can now read back a published artifact's contents; browsing, multi-file publishing and deletion are built but off. 2.1.239
CLAUDE_CODE_ARTIFACT_DIRECT_UPLOADArtifact availability now hinges on one gate, and direct upload kicks in for create paths. 2.1.172
CLAUDE_CODE_ARTIFACT_LIVE_BASE_URLThree artifact environment variables are now recognised, with the API token stripped from spawned processes as a secret. 2.1.221
Read out of the published npm bundle release by release, and out of Anthropic's own documentation as this site captured it. Nothing on this page is a description anybody here wrote about what the environment variable does. All environment variables.