Claude Code Review officially released, with review speed reportedly nearly 3 x faster

Article

Claude Code Review officially released: review speed nearly 3 x faster, “catches almost every bug”

Dongcha Beating AI News Flash: Anthropic has moved Claude Code's automated code review tool, Code Review, from research preview to general availability, opening it to Team and Enterprise users. The feature was launched as early as March this year.

After developers submit GitHub PR(code merge requests), it dispatches multiple Agent to inspect the code in parallel, looking for logic errors, security vulnerabilities, and edge cases, then filters out false positives through cross-validation. It ultimately places the issues, triggering conditions, and remediation suggestions directly on the relevant lines of code.

Anthropic says that most reviews now take only 6 to 10 minutes, nearly 20 minutes faster than the approximately 3 minutes required by the preview version. Administrators can also set three review-depth levels—Quick, Standard, and Deep—for different repositories.Claude Code founder Boris Cherny said that all of his PR use Code Review, and that it can detect nearly every bug before merging.

The feature requires an additional fee and is not included in Claude subscription quotas. Anthropic previously announced that each review costs an average of 15 to 25 USD, with actual charges based on Token usage. The more complex the code and the deeper the review, the higher the cost is usually.

Original link https://m.theblockbeats.info/flash/372320

Key points

Anthropic has moved Claude Code's automated code review tool, Code Review, from research preview to general availability.
Code Review is available to Team and Enterprise users.
developers submit GitHub PR; multiple Agent can inspect logic errors, security vulnerabilities, and edge cases in parallel, while filtering out false positives through cross-validation.
The final results place the issues, triggering conditions, and remediation suggestions on the relevant lines of code.
Most reviews now take 6 to 10 minutes; the preview version took approximately 20 minutes.
Administrators can set three review-depth levels—Quick, Standard, and Deep—for different repositories.
The feature requires an additional fee and is not included in Claude subscription quotas.
Anthropic previously announced that each review costs an average of 15 to 25 USD, with actual charges based on Token usage.
Claude Code Founder Boris Cherny said that all of his PR use Code Review, and that nearly every bug can be detected before merging.

AI-assisted interpretation

The following is analysis, separate from reported facts. Verify important claims independently.

Code Review automatically checks GitHub PR before code is merged and places detected issues and remediation suggestions on the relevant lines of code. It is now generally available, but requires an additional fee.

Why it matters to readers

It has advanced code review from research preview to general availability and offers different review depths; speed and cost vary with code complexity and review depth.

Beginners can understand it as a tool that automatically checks code after submission, but the automated review results still require human confirmation.

Risks and unknowns

The materials provided do not include Anthropic's original announcement.
“It can detect nearly every bug” is Boris Cherny's personal statement; the materials provide no independent accuracy or false-positive-rate data.
The materials do not specify the regions where the general-availability version is offered, the specific pricing rules, or actual user feedback.

Claude Code Review officially released, with review speed reportedly nearly 3 x faster

Article

AI-assisted interpretation

Why it matters to readers

Risks and unknowns

Related Developments

Related concepts

Keep following

Security Center