[DOCS] CLI reference missing `claude ultrareview` non-interactive subcommand

Documentation Type

Missing documentation (feature not documented)

Documentation Location

https://code.claude.com/docs/en/cli-reference

Section/Topic

CLI commands table and the ultrareview page's command examples for scripted/CI usage

Current Documentation

No documentation currently exists for the terminal subcommand claude ultrareview [target].

The CLI reference lists top-level claude subcommands such as:

claude remote-control claude setup-token

but does not list claude ultrareview.

The ultrareview page has a section titled:

Run ultrareview from the CLI

but its examples are interactive slash commands:

/ultrareview /ultrareview 1234

The headless/programmatic docs also currently say:

built-in commands are only available in interactive mode

What's Wrong or Missing?

Changelog v2.1.120 added claude ultrareview [target] as a non-interactive way to run /ultrareview from CI or scripts.

The current docs only explain the interactive slash-command workflow. They do not document:

• the claude ultrareview [target] terminal syntax
• what values [target] accepts
• how this differs from /ultrareview inside an interactive Claude Code session
• that it prints findings to stdout
• the --json raw-output mode
• exit-code behavior: 0 on completion and 1 on failure
• how to use it from CI or scripts

This also conflicts with the general headless guidance that built-in commands are only available in interactive mode, because claude ultrareview is now a specific non-interactive entry point for one built-in command.

Suggested Improvement

Add claude ultrareview [target] to the CLI commands table with a short description and example.

Suggested coverage:

• claude ultrareview reviews the current branch or default target, matching /ultrareview behavior where applicable
• claude ultrareview <target> documents the accepted target forms, such as a PR number if that is supported
• --json prints the raw review output for automation
• findings are printed to stdout in the default mode
• the command exits 0 when the review completes and 1 when it fails
• the command is intended for CI and scripts where an interactive slash command is not practical

Also update the ultrareview page's "Run ultrareview from the CLI" section to distinguish:

• interactive CLI slash command usage: /ultrareview
• terminal subcommand usage: claude ultrareview [target]

Finally, add a short exception or cross-reference in the headless/programmatic docs so the "built-in commands are only available in interactive mode" note does not make claude ultrareview look unsupported.

Impact

Medium - Makes feature difficult to understand

Additional Context

Affected Pages:

| Page | Line(s) | Context |
|------|---------|---------|
| https://code.claude.com/docs/en/cli-reference | 9-33 | Top-level CLI command table omits claude ultrareview |
| https://code.claude.com/docs/en/ultrareview | 23-45 | "Run ultrareview from the CLI" only shows interactive /ultrareview forms |
| https://code.claude.com/docs/en/headless | 23-30, 184-186 | Programmatic CLI docs describe -p usage and say built-in commands are only available in interactive mode |
| https://code.claude.com/docs/en/commands | 98 | Slash-command reference documents /ultrareview [PR], but not the terminal subcommand |

Total scope: 4 pages affected

Source: Changelog v2.1.120

Exact changelog entry:

Added claude ultrareview [target] subcommand to run /ultrareview non-interactively from CI or scripts — prints findings to stdout (--json for raw output) and exits 0 on completion or 1 on failure