What
/code-review ultragained new--post/--no-postflags that drive an end-to-end flow for posting an ultra code review's findings to a GitHub.com pull request as a single comment, via a new server route. A large set of specific, user-facing refusal or error messages cover cases like being rate-limited, not logged in, GitHub not connected, permission denied, or findings still pending.- When a cloud code review is configured to post its findings to the pull request from the user's own GitHub account, the initial "Cloud review completed" notification now notes that the posting is happening asynchronously and that a follow-up notification will report the outcome, and advises against applying the findings locally unless that was also requested.
Why
This lets a review's findings be shared directly on the pull request for others to see, without the user manually copying them over, while keeping the user informed about what's still in progress and what not to do while it finishes.
Feature flag
tengu_ultrareview_post_enabled Not enough to sayNothing here resolved what this flag was doing on this version, so nothing here should be read as on or off.
This account: no value returned · anonymous baseline: no value returned · compiled default in v2.1.269: on
Read once, for one account on one subscription tier, against v2.1.269. It isn't a statement about your account. What a flag value here can and cannot tell you
Documented inclaude-code/cli-reference
--post
CLI reference modified, high confidence
| `claude ultrareview [target]` | Run [ultrareview](/docs/en/ultrareview#run-ultrareview-non-interactively) non-interactively. Prints findings to stdout and exits 0 on success or 1 on failure. Use `--json` for the raw payload and `--timeou…see the edit
--no-post
CLI reference modified, high confidence
| `claude ultrareview [target]` | Run [ultrareview](/docs/en/ultrareview#run-ultrareview-non-interactively) non-interactively. Prints findings to stdout and exits 0 on success or 1 on failure. Use `--json` for the raw payload and `--timeou…see the edit
--post
Claude Code changelog modified, high confidence
* Changed `/ultrareview --post` to post the PR comment directly when the findings arrive and print the comment link, instead of starting a second cloud session to post itsee the edit
The entry above is what we published on the day. These lines were added later, as Anthropic's own pages caught up, and they sit beside the original rather than replacing it.
Added since
A small documentation edit on Code Review touched a line naming /code-review ultra after this was published.
Ultrareview requires authentication with a claude.ai account and is not available on Amazon Bedrock, Google Cloud's Agent Platform, or Microsoft Foundry, or to organizations with Zero Data Retention enabled or with the [HIPAA configuration…code-review see the edit
Confirmed since
Anthropic's documentation has since written up --no-post, on Find bugs with ultrareview.
| `--no-post` | Don't post the findings. This is the default, and if you pass both flags, Claude Code doesn't post. Requires Claude Code v2.1.227 or later |ultrareview see the edit
Confirmed since
Anthropic's documentation has since written up /code-review ultra, on Find bugs with ultrareview.
Ultrareview is a deep code review that runs as a [cloud session](/docs/en/claude-code-on-the-web) on Anthropic's infrastructure. When you run `/code-review ultra`, Claude Code launches a fleet of reviewer agents in a cloud sandbox to find …ultrareview see the edit
Confirmed since
Anthropic's documentation has since written up --post, on Claude Code changelog.
* \[Code Review] Fixed `/ultrareview --post` so a retry after a GitHub error posts the findings comment exactly once instead of never or twice; the comment now names the reviewed commitchangelog see the edit
Two sources agreeTwo things we can check say the same as this entry.
Anthropic's documentation agrees
--post on CLI reference
Anthropic's release notes agree
Changed /ultrareview --post to post the PR comment directly when the findings arrive and print the comment link, instead of starting a…