CodeRabbit Documentation - AI code reviews on pull requests, IDE, and CLI

Generated from the help output of CodeRabbit CLI 0.9.0. cr is the short alias for coderabbit. Run coderabbit <command> --help to see the same information in your terminal.

Global usage

CodeRabbit CLI for code review and cloud coding. Usage: coderabbit [options] [command]

Option	Description
`-v, --version`	Show CLI version

Mode	Description
`--agent`	Emit structured findings for agents
(default)	Plain-text review of tracked changes

Environment variables

Variable	Description
`CODERABBIT_HOST_URL`	Service base URL; `--host-url` `<url>` takes precedence
`CR_CLI_ALLOW_UNTRUSTED_HOST_URL`	Set true to allow a custom host; HTTPS required except loopback
`CR_CLI_TELEMETRY_DISABLED`	Set true, 1, yes, or on to disable telemetry
`CODERABBIT_CLI_DISABLE_AUTO_UPDATE`	Set true to disable automatic updates
`CODERABBIT_CLI_UPDATE_MANAGED`	Set true for organization-managed updates
`CODERABBIT_LOG_LEVEL`	debug, info, warn, or error (default: info)
`CODERABBIT_LOG_TRANSPORT`	console (stderr), file, or both (default: file)
`CR_CLI_LOG_FILE`	Custom log file path
`CR_CLI_AUTH_TIMEOUT`	Authentication HTTP timeout in milliseconds (default: 30000)
`CR_CLI_AGENT`	Agent name (e.g. codex, claude); can select noninteractive agent output

API keys use --api-key “<key>” with review or auth login; there is no API-key environment variable.

Commands

Command	Description
`auth`	Manage authentication
`review`	Review local changes
`pullrequest`	Inspect a CodeRabbit-reviewed GitHub pull request
`stats`	Show statistics from locally stored review history
`usage`	Show included-review availability and billing-period usage
`update`	Update CodeRabbit CLI to the latest version
`config`	Create or configure CodeRabbit configuration
`doctor`	Check CodeRabbit CLI installation and local review readiness
`skills`	Install or update CodeRabbit agent skills
`code`	Work with the CodeRabbit cloud Coding Agent
`completion`	Print a shell completion script

auth

Authentication commands Usage: coderabbit auth [options] [command]

Command	Description
`login [options]`	Authenticate with CodeRabbit
`logout [options]`	Clear local authentication
`status [options]`	Show current authentication status
`org [options]`	Switch the active organization

Examples

# Authenticate with CodeRabbit
coderabbit auth login

# Show current authentication status
coderabbit auth status

# Switch the active organization
coderabbit auth org

# Clear local authentication
coderabbit auth logout

Authenticate with CodeRabbit Usage: coderabbit auth login [options]

Option	Description
`--agent`	Emit structured JSON output for agent-driven OAuth login
`--self-hosted`	Use self-hosted CodeRabbit instance
`--no-browser`	Open the login URL yourself using a browser that can reach this machine’s localhost callback (port forwarding may be required)
`--api-key <api-key>`	Authenticate with a CodeRabbit API key
`--region <region>`	Use CodeRabbit’s US or EU region Choices: `us`, `eu`

Examples

# Interactive login with browser
coderabbit auth login

# Agent-friendly OAuth login with JSON output
coderabbit auth login --agent

# Manually open the URL with localhost access
coderabbit auth login --agent --no-browser

# Self-hosted instance login
coderabbit auth login --self-hosted

# Authenticate with CodeRabbit EU
coderabbit auth login --region eu

# Store an API key for CLI reviews
coderabbit auth login --api-key "<api-key>"

# Store an EU API key
coderabbit auth login --region eu --api-key "<api-key>"

--agent is only supported for localhost browser callbacks and never accepts pasted OAuth tokens.

auth logout

Clear local authentication Usage: coderabbit auth logout [options]

Option	Description
`--agent`	Emit structured JSON output for agent-driven logout

Examples

# Clear local authentication
coderabbit auth logout

# Emit structured logout output
coderabbit auth logout --agent

auth status

Show current authentication status Usage: coderabbit auth status [options]

Option	Description
`--agent`	Emit structured JSON output for agent-driven status check

Examples

# Show current authentication status
coderabbit auth status

# Emit structured auth status as JSON
coderabbit auth status --agent

auth org

Switch the active organization Usage: coderabbit auth org [options]

Option	Description
`--agent`	List available organizations as JSON without switching organizations

Examples

# Switch the active organization
coderabbit auth org

# Emit available organizations as JSON
coderabbit auth org --agent

review

AI-driven code review for the current Git workspace Usage: coderabbit review [options] [command] A review that fails or doesn’t cover every file exits with code 1, even after printing findings; see Exit codes. For which changes each scope flag selects, see Local review.

Option	Description
`--deep [focus]`	Use the full pull request review policy (focus text requires early access)
`--fresh`	Review selected changes without reusing the previous local checkpoint
`--show-prompts`	Print AI prompts without starting a new review
`--agent`	Emit structured findings for agent workflows
`--use-credits`	Use usage credits if this review exceeds included limits
`--committed`	Review only committed changes
`--uncommitted`	Review uncommitted Git changes
`--include-untracked`	Also review files that have not been added to Git
`-c, --config <files...>`	Additional instructions for CodeRabbit AI (e.g., claude.md, coderabbit.yaml)
`--base <branch>`	Git base branch for comparison
`--base-commit <commit>`	Git base commit on current branch for comparison
`--dir <path>`	Review only changes inside this directory
`--remote <owner/repo>`	Review a GitHub repository on the server without a local checkout
`--source-branch <ref>`	Source branch or ref to review with `--remote`
`--api-key <key>`	API key for authentication
`--usage`	Show included reviews and billing-period usage instead of reviewing (alias of `coderabbit usage`)
`--region <region>`	Region for inline API key authentication Choices: `us`, `eu`

Command	Description
`findings [options]`	Show the most recent local review with stored findings

Examples

# Review tracked changes
coderabbit review

# Review only committed changes
coderabbit review --committed

# Review uncommitted Git changes
coderabbit review --uncommitted

# Also review files not added to Git
coderabbit review --include-untracked

# Emit structured findings for agents
coderabbit review --agent

# Use the full pull request review policy
coderabbit review --deep

# Review selected changes without the previous local checkpoint
coderabbit review --fresh

# Print prompts from the latest local review
coderabbit review --show-prompts

# Compare current branch against main
coderabbit review --base main

# Review changes inside that directory
coderabbit review --dir /path/to/workspace

# Review with an EU API key without storing it
coderabbit review --region eu --api-key "<key>"

# Show stored local review findings
coderabbit review findings

# Forget stored findings for this scope
coderabbit review findings --clear

# Review a GitHub repository without a local clone
coderabbit review --remote acme/app --base main --source-branch my-feature

Plain text is the default review mode.
Use coderabbit auth login --agent for agent-driven OAuth login.
--remote requires --base and --source-branch, and runs outside any git checkout.

review findings

Show the most recent local review with stored findings Usage: coderabbit review findings [options]

Option	Description
`--dir <path>`	Review directory to read findings for
`--clear`	Dismiss stored findings for this scope so they stop being surfaced

Examples

# Show the most recent local review with stored findings
coderabbit review findings

# Show findings stored for a scoped review directory
coderabbit review findings --dir src

# Dismiss all stored findings for this scope
coderabbit review findings --clear

A newer review with no stored findings does not hide older findings.
Clearing is scoped to the review directory, current branch, and base branch.
Findings stored under any other scope are left unchanged.

pullrequest

Read existing CodeRabbit output without starting a review. Requires exactly one of --show-prompts or --show-threads. Usage: coderabbit pullrequest [options] <number-or-url> This command reads CodeRabbit’s existing output for the pull request and does not start a review. Pass exactly one of --show-prompts or --show-threads: the command fails without one, and the two cannot be combined. If no prompt is found, check that reviews.enable_prompt_for_ai_agents is enabled in the configuration and that CodeRabbit has finished reviewing the pull request. The command needs a CodeRabbit SaaS login or a stored API key, and self-hosted logins are not supported. Unlike coderabbit review, it does not start sign-in for you: if you have neither, run coderabbit auth login first.

Argument	Description
`<number-or-url>`	GitHub pull request number or URL

Option	Description
`--show-prompts`	Print the consolidated prompt for AI agents
`--show-threads`	Print CodeRabbit inline review-thread roots as JSON
`--agent`	Output newline-delimited JSON for agent integrations

Examples

coderabbit pullrequest https://github.com/org/repo/pull/123 --show-prompts
coderabbit pullrequest https://github.com/org/repo/pull/123 --show-prompts --agent

# Read current inline thread roots
coderabbit pullrequest 123 --show-threads --agent

# Resolve the repository from local origin
coderabbit pullrequest 123 --show-prompts

CodeRabbit SaaS sign-in or a stored agentic API key is required; the repository must be installed in the active organization.
Only github.com pull requests are supported. A full URL works outside a git checkout.

stats

Show statistics from locally stored review history Usage: coderabbit stats [options]

Option	Description
`--rebuild`	Rebuild stats by rescanning review history

Examples

# Show stats (builds on first run)
coderabbit stats

# Rescan review history and rebuild stats
coderabbit stats --rebuild

usage

Show included-review availability and billing-period usage Usage: coderabbit usage [options]

Option	Description
`--agent`	Emit newline-delimited JSON for agent workflows

Examples

# Show included reviews and billing-period usage
coderabbit usage

# Emit one machine-readable usage event
coderabbit usage --agent

# Same, via the default review command
coderabbit --usage

update

Update CodeRabbit CLI to the latest version Usage: coderabbit update [options] When a newer version is available and Homebrew manages your installation, coderabbit update installs nothing. It prints the new version and the brew upgrade coderabbit command, then exits with status 0, so a script can’t treat a successful exit as proof that the CLI was updated. On Windows, the update finishes after the command exits; see Update CodeRabbit CLI.

Examples

# Check for and install updates
coderabbit update

Homebrew installations must be updated with brew upgrade coderabbit.

config

Create or configure repository settings through a guided flow Usage: coderabbit config [options] [command]

Option	Description
`--agent`	Emit structured state for agent-guided setup without prompts
`--profile <profile>`	Review style for an agent-generated Standard proposal Choices: `chill`, `quiet`, `assertive`, `default`
`--generate`	Create or configure the repository settings (default)
`--detailed`	Manual setup: review style and path guidance (full setup via /config)

Examples

coderabbit config
coderabbit config --generate
coderabbit config --detailed
coderabbit config --agent
coderabbit config --agent --generate

Schema: https://www.coderabbit.ai/integrations/schema.v2.json

doctor

Check CodeRabbit CLI installation and local review readiness Usage: coderabbit doctor [options]

skills

Install or update every skill in the latest verified CodeRabbit skills release. Detected agents preview changes for approval; terminal users confirm interactively. Usage: coderabbit skills [options]

Option	Description
`--agent`	Preview changes for approval in an agent conversation
`--confirm <plan-hash>`	Apply the exact previewed plan after user approval, without prompting

Examples

coderabbit skills
coderabbit skills --agent

Agent flow: show the plan to the user and ask for approval. Only after approval, run the command in the final action_required JSON record. A changed plan requires fresh approval. No skill files are changed by the preview.

code

Use the same Coding Agent and skill library as the web app. Coding Agent commands are in beta. Usage: coderabbit code [options] [command] Every coderabbit code command needs a CodeRabbit SaaS user login and acts in your selected organization. API key, self-hosted, and SSO workspace logins are not supported (requirements). If no organization is selected, run coderabbit auth org in an interactive terminal first. To learn what the cloud Coding Agent does with the work you send it, see Coding Agent.

Command	Description
`handoff [options]`	Continue a local coding session in a new cloud task
`ls [options]`	List your cloud Coding Agent tasks
`show [options] <task>`	Show a cloud Coding Agent task
`new [options] [prompt]`	Start a cloud Coding Agent task
`resume [options] [task]`	Follow and continue a cloud Coding Agent task
`cancel [options] <task>`	Stop the running turn of a cloud Coding Agent task
`ask [options] <task> <question>`	Ask a side question about a cloud Coding Agent task
`plan [options] <task>`	Show or approve the plan of a cloud Coding Agent task
`push [options] <task>`	Push the changes of a cloud Coding Agent task
`autopilot [options] <action> [task]`	Control Autopilot for a cloud task or pull request
`skills`	Manage cloud Coding Agent skills

code handoff

Create a new CodeRabbit Cloud task with a summary, optional plan, and automatically discovered session transcript Usage: coderabbit code handoff [options] --summary is required. Run the command from a clean checkout of a branch that has an origin remote. Untracked files count as changes, so keep the summary and plan files outside the repository or commit them first. For what is uploaded and what happens next, see Continue a local session in the cloud.

Option	Description
`--summary <path\|->`	Session summary file, or - to read from stdin (required)
`--plan <path>`	Primary implementation plan file
`--agent`	Emit newline-delimited JSON for agent workflows

code ls

List the 20 most recently updated cloud tasks in the selected organization. Only your tasks are listed unless you pass --all. Archived tasks are not listed. Usage: coderabbit code ls [options]

Option	Description
`--status <state>`	Only list tasks in this web task list state Choices: `needs_attention`, `ready_for_review`, `running`, `completed`, `canceled`
`--repo <path>`	Only list tasks for the Git repository at this path, for example .
`--all`	Include tasks from other people in the organization
`--agent`	Emit newline-delimited JSON for agent workflows

code show

Show the status, repository, delivery, and pull request health of a cloud task. Usage: coderabbit code show [options] <task>

Argument	Description
`<task>`	Task ID or task URL

Option	Description
`--agent`	Emit newline-delimited JSON for agent workflows

code new

Start a cloud task on the current branch of the Git repository in this directory. The task starts from that branch on origin. Local commits that are not pushed and uncommitted changes are not included. Usage: coderabbit code new [options] [prompt]

Argument	Description
`[prompt]`	What the agent should do, or - to read it from stdin. Interactive mode prompts when omitted; agent mode requires it

Option	Description
`--plan`	Start in Plan mode: the agent writes a plan for review instead of changing code
`--resume`	Follow the task after it starts, as `coderabbit code resume` does
`--agent`	Emit newline-delimited JSON for agent workflows

code resume

Open a cloud task in the terminal. Shows its trace and live updates, and lets you send messages, steer the running turn, answer agent questions, and stop the turn. Closing the view does not stop the task. With --agent, send at most one message, steer, or answer, then stream newline-delimited JSON until the turn ends, the agent asks a blocking question, or 9 minutes pass. Usage: coderabbit code resume [options] [task]

Argument	Description
`[task]`	Task ID or task URL. In a terminal, omit it to pick one of your recent tasks; with `--agent`, pass it or `--last`

Option	Description
`--last`	Use your most recently updated task for the Git repository in the current directory
`-m, --message <message>`	With `--agent`: send a message. Use - to read it from stdin
`--steer <message>`	With `--agent`: steer the running turn. Use - to read it from stdin
`--answer <json>`	With `--agent`: answer the pending question with the JSON from its answerTemplate. Use - to read it from stdin
`--agent`	Emit newline-delimited JSON for agent workflows

code cancel

Stop the running turn and drop queued messages. The task stays open: a new message starts a new turn. Usage: coderabbit code cancel [options] <task>

Argument	Description
`<task>`	Task ID or task URL

Option	Description
`--agent`	Emit newline-delimited JSON for agent workflows

code ask

Ask the task’s side chat a question and print the answer. A side chat reads the task’s conversation and repository. It does not change the task, and the turn of the task keeps running. Each question is billed. The side chat stays open for follow-up questions: pass --follow-up with the turn ID of an earlier answer to keep its context. A task can have 3 open side chats. An idle side chat expires after 24 hours. The CLI cannot close a side chat, so ask later questions about the task with --follow-up. The command waits up to 3 minutes for the task’s environment and up to 15 minutes for the answer. Usage: coderabbit code ask [options] <task> <question>

Argument	Description
`<task>`	Task ID or task URL
`<question>`	The question. Use - to read it from stdin

Option	Description
`--follow-up <turn-id>`	Ask in the side chat of an earlier answer. Pass the turn ID that the answer printed
`--agent`	Emit newline-delimited JSON for agent workflows

code plan

Show the latest finalized plan of a cloud task. With --approve, record your approval of that plan version. You cannot approve the plan of a task that you started, and approval does not start implementation. Usage: coderabbit code plan [options] <task>

Argument	Description
`<task>`	Task ID or task URL

Option	Description
`--approve`	Approve the latest plan version as a reviewer
`--agent`	Emit newline-delimited JSON for agent workflows

code push

Deliver the ready changes of a cloud task and wait until the delivery finishes. By default, CodeRabbit commits and pushes the changes, as the web app’s Commit and push action does. GitHub can first ask you to authorize CodeRabbit in a browser that is signed in to CodeRabbit. Usage: coderabbit code push [options] <task>

Argument	Description
`<task>`	Task ID or task URL

Option	Description
`--stacked`	Open a new pull request with the changes, as the web app’s Create PR action does
`--agent`	Emit newline-delimited JSON for agent workflows

code autopilot

Autopilot publishes the task’s changes when the agent finishes, then fixes CodeRabbit findings, required CI, and merge conflicts. on starts it, off stops it, resume starts a new repair episode after a pause, and status shows its state. Pass exactly one target: a task ID or URL, or --pr <number|url>. on and resume check Coding Agent billing; with --pr they can create an Autopilot task for the pull request. Usage: coderabbit code autopilot [options] <action> [task]

Argument	Description
`<action>`	on, off, resume, or status
`[task]`	Task ID or task URL

Option	Description
`--pr <number\|url>`	Target a pull request: a number in the current repository, or a GitHub pull request or GitLab merge request URL
`--agent`	Emit newline-delimited JSON for agent workflows

code skills

Manage cloud Coding Agent skills Usage: coderabbit code skills [options] [command]

Command	Description
`import [options] [path]`	Import local agent skills into your cloud library

code skills import

Choose an agent, select local skills, and review their import into the cloud library. Pass a skill folder or SKILL.md to import one directly. With --yes and no --scope, new skills are personal within the selected organization and existing skills keep their visibility. Importing an owned skill with the same name publishes a new version. Noninteractive imports require a path and --yes. Usage: coderabbit code skills import [options] [path]

Argument	Description
`[path]`	Skill folder or its SKILL.md

Option	Description
`--scope <scope>`	Who can use the imported skill Choices: `personal`, `organization`
`--yes`	Upload the specified path without prompting

completion

Print the shell completion script for coderabbit and cr. It completes commands, options, and the task IDs that this computer listed or opened. Usage: coderabbit completion [options] <shell>

Argument	Description
`<shell>`	Shell type for completion script Choices: `zsh`, `bash`, `fish`, `powershell`

Examples

# Add to ~/.bashrc
eval "$(coderabbit completion bash)"

# Add to ~/.zshrc, after compinit
eval "$(coderabbit completion zsh)"

# Add to ~/.config/fish/config.fish
coderabbit completion fish | source

# Add to $PROFILE
coderabbit completion powershell | Out-String | Invoke-Expression

The script completes coderabbit and the cr alias.
Task IDs come from the tasks that this computer listed or opened with coderabbit code. A Tab press makes no network call.

Command-Line Review Tool

IDE Extension

CLI Command Reference

Global usage

Environment variables

Commands

auth

auth logout

auth status

auth org

review

review findings

pullrequest

stats

usage

update

config

doctor

skills

code

code handoff

code ls

code show

code new

code resume

code cancel

code ask

code plan

code push

code autopilot

code skills

code skills import

completion

​Global usage

​Environment variables

​Commands

​auth

​auth login

​auth logout

​auth status

​auth org

​review

​review findings

​pullrequest

​stats

​usage

​update

​config

​doctor

​skills

​code

​code handoff

​code ls

​code show

​code new

​code resume

​code cancel

​code ask

​code plan

​code push

​code autopilot

​code skills

​code skills import

​completion

Global usage

Environment variables

Commands

auth

auth login

auth logout

auth status

auth org

review

review findings

pullrequest

stats

usage

update

config

doctor

skills

code

code handoff

code ls

code show

code new

code resume

code cancel

code ask

code plan

code push

code autopilot

code skills

code skills import

completion