# CodeRabbit

AI-powered code reviews and software development insights from CodeRabbit

This is one page of public article previews, not the complete archive. Follow Next page to continue. Summaries are not the original full articles.

## AI pull requests in open source have a context problem

DevFeed: [AI pull requests in open source have a context problem](<https://devfeed.tech/articles/ai-pull-requests-in-open-source-have-a-context-problem-144039.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/ai-pull-requests-open-source-context-problem>)

Author: Hendrik Krack

Published: 2026-11-19T18:18:57.960894Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai slop pull requests](<https://devfeed.tech/topics/ai-slop-pull-requests.md>), [Maintainers](<https://devfeed.tech/topics/maintainers.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [context](<https://devfeed.tech/topics/context.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [context](<https://devfeed.tech/tags/context.md>), [contribution](<https://devfeed.tech/tags/contribution.md>), [english](<https://devfeed.tech/tags/english.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [prompts](<https://devfeed.tech/tags/prompts.md>), [pull-requests](<https://devfeed.tech/tags/pull-requests.md>)

### AI overview

Agent-written pull requests often lack project context, leading to changes that violate contribution rules or duplicate existing fixes and leaving maintainers to reconstruct the rationale. The excerpt describes responses including placing instructions beside the code in AGENTS.md and CLAUDE.md, rejecting PRs that ignore agent contribution guides, publishing AI policies, and limiting how many PRs outside contributors can open. These measures do not remove the need for human approval: prompts.chat uses CodeRabbit and sometimes Copilot for review, but Fatih Kadir Akın still reads changes before merging. He advises contributors to give agents project context and build verification skills; the excerpt notes that agents can still make basic mistakes, such as introducing Ruby into a JavaScript project. Maintainers must weigh the review burden against the possibility that AI-generated reports uncover real flaws.

### Source excerpt

Agent-written PRs keep ignoring contribution guides. Here's what prompts.chat creator Fatih Kadir Akın thinks agents need, starting with context.

## Introducing CodeRabbit Verify: Agentic application testing and AI quality assurance

DevFeed: [Introducing CodeRabbit Verify: Agentic application testing and AI quality assurance](<https://devfeed.tech/articles/introducing-coderabbit-verify-agentic-application-testing-and-ai-quality-assurance-140985.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/introducing-coderabbit-verify>)

Author: Tejas Venugopal

Published: 2026-11-18T12:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Test automation](<https://devfeed.tech/topics/test-automation.md>), [ui-testing](<https://devfeed.tech/topics/ui-testing.md>), [Integration testing](<https://devfeed.tech/topics/integration-testing.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [bugfixes](<https://devfeed.tech/topics/bugfixes.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-quality](<https://devfeed.tech/tags/ai-quality.md>), [application](<https://devfeed.tech/tags/application.md>), [automatically](<https://devfeed.tech/tags/automatically.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [code](<https://devfeed.tech/tags/code.md>), [english](<https://devfeed.tech/tags/english.md>), [faster](<https://devfeed.tech/tags/faster.md>), [featured](<https://devfeed.tech/tags/featured.md>), [functional](<https://devfeed.tech/tags/functional.md>), [less](<https://devfeed.tech/tags/less.md>), [product](<https://devfeed.tech/tags/product.md>), [qa](<https://devfeed.tech/tags/qa.md>), [quality](<https://devfeed.tech/tags/quality.md>), [quality-assurance](<https://devfeed.tech/tags/quality-assurance.md>), [ship](<https://devfeed.tech/tags/ship.md>), [software](<https://devfeed.tech/tags/software.md>), [test](<https://devfeed.tech/tags/test.md>), [testing](<https://devfeed.tech/tags/testing.md>), [verify](<https://devfeed.tech/tags/verify.md>), [with](<https://devfeed.tech/tags/with.md>)

### AI overview

CodeRabbit Verify tests a running application against PR changes and reusable user journeys, aiming to catch functional bugs and broken workflows that unit tests or manual review may miss. For each PR, it builds editable scenarios from commits, source code, user-provided tests and documentation; existing CodeRabbit review context may also inform them. Expected behavior comes from intended behavior, docs and existing test expectations. Results are tied to the tested commit, and failures include observations and evidence that can be handed to CodeRabbit Agent or another coding agent. For regression testing, Verify proposes plain-English journeys for team review; approved journeys can be run on demand, tagged, or scheduled against a branch and test environment. Scheduled runs can create Jira or Linear issues. Verify records browser activity, screenshots and logs, and distinguishes application bugs from checks blocked by setup or other limitations. CodeRabbit Agent prepares and snapshots a sandbox environment for reuse, separate from production. The excerpt says PR runs, full-repository journey runs and CodeRabbit Agent fixes are free during the public beta; PR testing starts with `@ไ`

### Source excerpt

Automatically test your agentic code, catch functional bugs, and ship software faster with less manual QA

## AI feature bloat: why shipping less is better

DevFeed: [AI feature bloat: why shipping less is better](<https://devfeed.tech/articles/ai-feature-bloat-why-shipping-less-is-better-140984.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/ai-feature-bloat-why-shipping-less-is-better>)

Author: Hendrik Krack

Published: 2026-11-17T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [feature bloat](<https://devfeed.tech/topics/feature-bloat.md>), [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [User interface design](<https://devfeed.tech/topics/ui-design.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [ai-slop](<https://devfeed.tech/tags/ai-slop.md>), [api](<https://devfeed.tech/tags/api.md>), [cli](<https://devfeed.tech/tags/cli.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [cost](<https://devfeed.tech/tags/cost.md>), [english](<https://devfeed.tech/tags/english.md>), [feature-bloat](<https://devfeed.tech/tags/feature-bloat.md>), [feedback](<https://devfeed.tech/tags/feedback.md>), [focus](<https://devfeed.tech/tags/focus.md>), [gate](<https://devfeed.tech/tags/gate.md>), [shipping](<https://devfeed.tech/tags/shipping.md>)

### AI overview

Hassan El-Mghari argues that AI makes features cheap to build but not cheap to get wrong: half-working features can erode user trust. He tests demand by teasing apps before committing to a build, uses repeated requests in responses to prioritize features, and launches publicly to hear feedback, including reports of security flaws such as an unauthenticated API route. His scope test is whether an app can be described in five words, with emphasis on making one thing work well. The excerpt presents these as his practices and examples, not as evidence that they guarantee success.

### Source excerpt

Coding agents make adding features nearly free, and Hassan El-Mghari of Together AI argues that is exactly why you should ship fewer of them. He explains how he tests ideas before building anything and keeps every app describable in five words.

## The hard part of AI coding is deciding what to keep

DevFeed: [The hard part of AI coding is deciding what to keep](<https://devfeed.tech/articles/the-hard-part-of-ai-coding-is-deciding-what-to-keep-135978.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/the-merge-ryan-vogel-deciding-what-to-keep>)

Author: Hendrik Krack

Published: 2026-11-16T18:12:39.425207Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [codex](<https://devfeed.tech/topics/codex.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [api](<https://devfeed.tech/tags/api.md>), [app](<https://devfeed.tech/tags/app.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [bug](<https://devfeed.tech/tags/bug.md>), [business-logic](<https://devfeed.tech/tags/business-logic.md>), [caching](<https://devfeed.tech/tags/caching.md>), [challenge](<https://devfeed.tech/tags/challenge.md>), [code-generation](<https://devfeed.tech/tags/code-generation.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [coding](<https://devfeed.tech/tags/coding.md>), [collaboration](<https://devfeed.tech/tags/collaboration.md>), [comment](<https://devfeed.tech/tags/comment.md>), [complexity](<https://devfeed.tech/tags/complexity.md>), [engineering](<https://devfeed.tech/tags/engineering.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [pr](<https://devfeed.tech/tags/pr.md>), [review](<https://devfeed.tech/tags/review.md>)

### AI overview

The harder part of AI-written code is deciding whether a change belongs, not merely reviewing whether it passes. Ryan Vogel recommends stating the app's needs, challenging unnecessary architecture, and using a separate agent to check intermediate work against the original direction. For code review, Harjot Gill says CodeRabbit reconstructs change order and reviews from data and schema through logic and APIs to interface and tests, helping reviewers trace dependencies. The discussion also favors testable evidence over confident claims: an agent should be able to ask questions or run checks, and suspected bugs should be tested before becoming findings. Clear-end tasks suit agents better; open-ended design needs more frequent human decisions. Handoffs should expose decisions, dependencies, evidence, and remaining human judgment, with review effort scaled to a change's consequences.

### Source excerpt

A PR can pass review while nobody decides if the change belongs. Harjot Gill and Ryan Vogel discuss engineering judgment in agent-written code.

## CodeRabbit went looking for trouble in open-source code and isn't done yet

DevFeed: [CodeRabbit went looking for trouble in open-source code and isn't done yet](<https://devfeed.tech/articles/coderabbit-went-looking-for-trouble-in-open-source-code-and-isn-t-done-yet-132557.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-secure-deep-scan-open-source>)

Author: Santosh Yadav

Published: 2026-11-13T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [software composition analysis](<https://devfeed.tech/topics/software-composition-analysis.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [open source maintenance](<https://devfeed.tech/topics/open-source-maintenance.md>), [snyk-open-source](<https://devfeed.tech/topics/snyk-open-source.md>)

Tags: [access-controls](<https://devfeed.tech/tags/access-controls.md>), [accounting](<https://devfeed.tech/tags/accounting.md>), [actions](<https://devfeed.tech/tags/actions.md>), [can](<https://devfeed.tech/tags/can.md>), [code](<https://devfeed.tech/tags/code.md>), [critical](<https://devfeed.tech/tags/critical.md>), [data](<https://devfeed.tech/tags/data.md>), [deep](<https://devfeed.tech/tags/deep.md>), [english](<https://devfeed.tech/tags/english.md>), [exposed](<https://devfeed.tech/tags/exposed.md>), [featured](<https://devfeed.tech/tags/featured.md>), [first](<https://devfeed.tech/tags/first.md>), [free](<https://devfeed.tech/tags/free.md>), [net](<https://devfeed.tech/tags/net.md>), [open](<https://devfeed.tech/tags/open.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-source-projects](<https://devfeed.tech/tags/open-source-projects.md>), [projects](<https://devfeed.tech/tags/projects.md>), [public](<https://devfeed.tech/tags/public.md>), [repositories](<https://devfeed.tech/tags/repositories.md>), [scan](<https://devfeed.tech/tags/scan.md>), [security](<https://devfeed.tech/tags/security.md>), [source](<https://devfeed.tech/tags/source.md>)

### AI overview

CodeRabbit Secure Deep Scans examined 1,125 public repositories from October 23 to November 3 and reported 1,215 high or critical findings (1,200 high, 15 critical); at least 75 had been fixed. The scan maps entry points and trust boundaries, follows code paths to investigate suspected issues, and uses a separate verifier to check reachability and safeguards. For supported findings, Fix with AI prepares a patch and opens a pull request for maintainer review. Deep Scan is presented as a complement to static analysis and dependency scanning, aimed at application-specific flaws those configured checks may miss. Open-source projects trying it for the first time can get one free scan, up to $1,000; projects need at least two members to qualify.

### Source excerpt

Missing access checks and exposed data were among the 1,215 high or critical findings CodeRabbit Secure uncovered across 1,125 public repositories. Now, eligible open-source projects can get their first Deep Scan free.

## CodeRabbit Passes $100 million in ARR

DevFeed: [CodeRabbit Passes $100 million in ARR](<https://devfeed.tech/articles/coderabbit-passes-100-million-in-arr-130674.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-100-million-arr>)

Author: Harjot Gill

Published: 2026-11-12T12:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [code security](<https://devfeed.tech/topics/code-security.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [3](<https://devfeed.tech/tags/3.md>), [about](<https://devfeed.tech/tags/about.md>), [added](<https://devfeed.tech/tags/added.md>), [administration](<https://devfeed.tech/tags/administration.md>), [adoption](<https://devfeed.tech/tags/adoption.md>), [after](<https://devfeed.tech/tags/after.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-loop](<https://devfeed.tech/tags/agentic-loop.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-adoption](<https://devfeed.tech/tags/ai-adoption.md>), [ai-generated-code](<https://devfeed.tech/tags/ai-generated-code.md>), [ai-native](<https://devfeed.tech/tags/ai-native.md>), [ai-powered](<https://devfeed.tech/tags/ai-powered.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [annual-recurring-revenue](<https://devfeed.tech/tags/annual-recurring-revenue.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [just](<https://devfeed.tech/tags/just.md>), [over](<https://devfeed.tech/tags/over.md>), [revenue](<https://devfeed.tech/tags/revenue.md>)

### AI overview

CodeRabbit says its annual recurring revenue grew from $10 million to $100 million in just over a year. It reports more than 20,000 customers, 400,000 monthly active users, and over 3 million pull requests reviewed weekly; the average team ships at twice its previous rate. The product provides an independent AI review layer that assesses changes for issues, standards, architectural fit, and risk, with monitoring after merge. The company argues this helps teams validate agent-written code as quickly as it is generated, while keeping review separate from the code-writing agent. It has added OpenAI GPT Cyber models to CodeRabbit Secure and reports that an initial free-scan program found over 43,000 critical or high-severity findings across 13,000 scans. CodeRabbit is also available through several AI providers' marketplaces and says its review pipeline is model-agnostic.

### Source excerpt

CodeRabbit has grown from $10 million to $100 million in annual recurring revenue in just over a year.

## Agentic software development with CodeRabbit and Google Antigravity

DevFeed: [Agentic software development with CodeRabbit and Google Antigravity](<https://devfeed.tech/articles/agentic-software-development-with-coderabbit-and-google-antigravity-130673.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/agentic-software-development-coderabbit-google-antigravity>)

Author: Sahil Bansal

Published: 2026-11-11T12:00:00Z

Content type: tutorial

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [google-antigravity](<https://devfeed.tech/topics/google-antigravity.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [Loop Engineering](<https://devfeed.tech/topics/loop-engineering.md>), [AI tools for developers](<https://devfeed.tech/topics/ai-tools-for-developers.md>), [Agent Skills](<https://devfeed.tech/topics/agent-skills.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [adversarial](<https://devfeed.tech/tags/adversarial.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-loop](<https://devfeed.tech/tags/agentic-loop.md>), [agentic-software](<https://devfeed.tech/tags/agentic-software.md>), [agentic-software-development](<https://devfeed.tech/tags/agentic-software-development.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [ai-code](<https://devfeed.tech/tags/ai-code.md>), [antigravity](<https://devfeed.tech/tags/antigravity.md>), [code](<https://devfeed.tech/tags/code.md>), [code-reviews](<https://devfeed.tech/tags/code-reviews.md>), [development](<https://devfeed.tech/tags/development.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [google](<https://devfeed.tech/tags/google.md>), [google-antigravity](<https://devfeed.tech/tags/google-antigravity.md>), [product](<https://devfeed.tech/tags/product.md>), [review](<https://devfeed.tech/tags/review.md>), [reviews](<https://devfeed.tech/tags/reviews.md>), [software](<https://devfeed.tech/tags/software.md>), [software-development](<https://devfeed.tech/tags/software-development.md>)

### AI overview

Google Antigravity can use CodeRabbit to review local code changes, receive contextual findings, apply fixes, and request another review in one agentic loop. Install the CodeRabbit plugin from Antigravity's Plugin Marketplace or with `agy plugin install coderabbit`; the separate CodeRabbit CLI performs local reviews and must be installed before the first review. Sign in with `coderabbit auth login`. The plugin's review skill guides Antigravity through fixing findings and reviewing again. Its Autofix skill applies existing GitHub pull request feedback, with approval for each change, and requires the GitHub CLI installed and signed in. The excerpt says CodeRabbit reviews for functional and logical errors, race conditions, security vulnerabilities, test coverage, and architectural concerns using codebase context and team guidelines; it does not provide comparative performance results.

### Source excerpt

Bring CodeRabbit's specialized AI code reviews into Google Antigravity for a build -> review -> fix agentic loop

## Multi-Repository Security Analysis with CodeRabbit Secure

DevFeed: [Multi-Repository Security Analysis with CodeRabbit Secure](<https://devfeed.tech/articles/multi-repository-security-analysis-with-coderabbit-secure-126472.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/multi-repository-security-analysis-with-coderabbit-secure>)

Author: Konrad Sopala

Published: 2026-11-10T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [code security](<https://devfeed.tech/topics/code-security.md>), [Static code analysis](<https://devfeed.tech/topics/static-code-analysis.md>), [Multi Agent Systems](<https://devfeed.tech/topics/multi-agent-systems.md>), [Secure token management](<https://devfeed.tech/topics/secure-token-management.md>), [Repository](<https://devfeed.tech/topics/repository.md>)

Tags: [agents](<https://devfeed.tech/tags/agents.md>), [api](<https://devfeed.tech/tags/api.md>), [api-gateway](<https://devfeed.tech/tags/api-gateway.md>), [application](<https://devfeed.tech/tags/application.md>), [architecture](<https://devfeed.tech/tags/architecture.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [bootstrap](<https://devfeed.tech/tags/bootstrap.md>), [branch](<https://devfeed.tech/tags/branch.md>), [bypass](<https://devfeed.tech/tags/bypass.md>), [charts](<https://devfeed.tech/tags/charts.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [context](<https://devfeed.tech/tags/context.md>), [credentials](<https://devfeed.tech/tags/credentials.md>), [deployment](<https://devfeed.tech/tags/deployment.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [repositories](<https://devfeed.tech/tags/repositories.md>), [repository](<https://devfeed.tech/tags/repository.md>), [scan](<https://devfeed.tech/tags/scan.md>), [security](<https://devfeed.tech/tags/security.md>), [security-analysis](<https://devfeed.tech/tags/security-analysis.md>)

### AI overview

CodeRabbit Secure's Deep Scan can use linked repositories as read-only context to trace settings, shared code, and deployment configuration while investigating findings in a primary repository. Architecture summaries and auto-linking help identify related repositories; teams can also configure links and source access. Agents use this context across recon, hunt, and verify, but linked repositories receive no findings or coverage from that scan and require separate scans. In two open-source experiments, linked startup code clarified that default administrator credentials applied during fresh installation, while a runtime check showed changing the bootstrap password did not rotate an existing account's password. In another experiment, supplying the runtime repository let the evaluator distinguish enabled from disabled reload configurations; the chart alone left both unresolved. Linked source is pinned to the default branch commit, which may differ from deployed code. Manual mode reads selected repositories; automatic mode reads eligible linked repositories up to the plan limit until the source list is customized. Supported platforms listed are GitHub Cloud, GitHub Enterprise Server, 

### Source excerpt

An application can have an authentication check and still expose data if a caller can bypass the service that enforces it. Multi-repository analysis in CodeRabbit Secure gives Deep Scan agents read-only context from linked repositories, so they can follow a configuration value, shared library, or deployment setting to the code that consumes it while investigating a finding.

## GPT-6 Cyber Security Evaluation: CodeRabbit's Results

DevFeed: [GPT-6 Cyber Security Evaluation: CodeRabbit's Results](<https://devfeed.tech/articles/gpt-6-cyber-security-evaluation-coderabbit-s-results-121126.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/gpt-6-cyber-security-evaluation>)

Author: Rohit Khanna

Published: 2026-11-05T12:00:00Z

Content type: news

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [AI in cybersecurity](<https://devfeed.tech/topics/ai-in-cybersecurity.md>), [Static code analysis](<https://devfeed.tech/topics/static-code-analysis.md>), [Security](<https://devfeed.tech/topics/security.md>)

Tags: [authorization](<https://devfeed.tech/tags/authorization.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [bypass](<https://devfeed.tech/tags/bypass.md>), [c-sharp](<https://devfeed.tech/tags/c-sharp.md>), [cases](<https://devfeed.tech/tags/cases.md>), [cross-site-scripting](<https://devfeed.tech/tags/cross-site-scripting.md>), [cyber-security](<https://devfeed.tech/tags/cyber-security.md>), [cybersecurity](<https://devfeed.tech/tags/cybersecurity.md>), [daybreak](<https://devfeed.tech/tags/daybreak.md>), [daybreak-defense-network](<https://devfeed.tech/tags/daybreak-defense-network.md>), [deep](<https://devfeed.tech/tags/deep.md>), [denial-of-service](<https://devfeed.tech/tags/denial-of-service.md>), [english](<https://devfeed.tech/tags/english.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [exploited](<https://devfeed.tech/tags/exploited.md>), [featured](<https://devfeed.tech/tags/featured.md>), [go](<https://devfeed.tech/tags/go.md>), [gpt-6-cyber](<https://devfeed.tech/tags/gpt-6-cyber.md>), [idor](<https://devfeed.tech/tags/idor.md>), [model-evaluation](<https://devfeed.tech/tags/model-evaluation.md>), [results](<https://devfeed.tech/tags/results.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

In CodeRabbit Secure's comparison on the same 33 cases, adding GPT-6 Cyber increased known-issue catches from 19 to 22 (a 15.8% relative gain) and raised high- and medium-severity findings in final reports by 65.6%, rounded to 66%. The net gain reflects five additional catches and two baseline catches lost; nine selected issues remained undetected by both configurations. Cases covered nine programming languages, and only issues present in final reports with an explanation received catch credit. The extra findings have not yet been assessed for validity, so report accuracy and prioritization remain open questions. Investigation used high reasoning effort, but no lower-effort run was compared, leaving its effect on coverage, time, and cost unknown.

### Source excerpt

Across 33 benchmark cases, CodeRabbit's ensemble with GPT-6 Cyber caught 22 known issues versus 19 without it and reported 66% more high- and medium-severity findings.

## How Runtime Verification works in CodeRabbit Secure

DevFeed: [How Runtime Verification works in CodeRabbit Secure](<https://devfeed.tech/articles/how-runtime-verification-works-in-coderabbit-secure-116314.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/runtime-verification-in-coderabbit-secure>)

Author: Konrad Sopala

Published: 2026-11-03T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Test automation](<https://devfeed.tech/topics/test-automation.md>), [Integration testing](<https://devfeed.tech/topics/integration-testing.md>)

Tags: [403](<https://devfeed.tech/tags/403.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [application](<https://devfeed.tech/tags/application.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [authentication](<https://devfeed.tech/tags/authentication.md>), [build](<https://devfeed.tech/tags/build.md>), [configuration](<https://devfeed.tech/tags/configuration.md>), [dependencies](<https://devfeed.tech/tags/dependencies.md>), [details](<https://devfeed.tech/tags/details.md>), [developer](<https://devfeed.tech/tags/developer.md>), [developers](<https://devfeed.tech/tags/developers.md>), [english](<https://devfeed.tech/tags/english.md>), [environment](<https://devfeed.tech/tags/environment.md>), [execution](<https://devfeed.tech/tags/execution.md>), [featured](<https://devfeed.tech/tags/featured.md>), [guide](<https://devfeed.tech/tags/guide.md>), [request](<https://devfeed.tech/tags/request.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [scan](<https://devfeed.tech/tags/scan.md>), [security](<https://devfeed.tech/tags/security.md>), [testing](<https://devfeed.tech/tags/testing.md>), [verification](<https://devfeed.tech/tags/verification.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

Runtime Verification turns a Deep Scan finding into a reviewable Reproducer Guide, then can run that guide with Runtime Analysis in a prepared CodeRabbit coding environment. CodeRabbit rechecks whether the issue is present before generating the guide; if the issue is absent or the recheck is inconclusive, guide generation stops. The guide specifies prerequisites, steps and expected observations, and pins the tested revision. Running it does not recheck the finding or test the latest source. Results include evidence and indicators for application readiness and whether the intended code path was exercised, and report Reproduced, Not reproduced or Inconclusive for the tested revision and environment. "Not reproduced" does not establish safety; setup failures or requests that miss the relevant code path may make the result inconclusive. Teams may inspect and run guides themselves, or enable automatic verification for new findings of a chosen severity after scans. Runtime Analysis requires a prepared environment, and the article advises using a dedicated test environment and avoiding production credentials.

### Source excerpt

Testing a suspected vulnerability takes more than sending a request and checking the response. Runtime Verification in CodeRabbit Secure turns a Deep Scan finding into a Reproducer Guide, runs it with Runtime Analysis in a prepared environment, and shows whether the application was ready and the intended code path was exercised.

## Why it's time to fight AI with AI

DevFeed: [Why it's time to fight AI with AI](<https://devfeed.tech/articles/why-it-s-time-to-fight-ai-with-ai-105810.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/fighting-ai-with-ai>)

Author: Konrad Sopala

Published: 2026-10-28T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Artificial Intelligence](<https://devfeed.tech/topics/ai.md>), [llm-reasoning](<https://devfeed.tech/topics/llm-reasoning.md>), [reproducibility](<https://devfeed.tech/topics/reproducibility.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [access-control](<https://devfeed.tech/tags/access-control.md>), [ai](<https://devfeed.tech/tags/ai.md>), [application-security](<https://devfeed.tech/tags/application-security.md>), [broken-access-control](<https://devfeed.tech/tags/broken-access-control.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

Deep Scan in CodeRabbit Secure investigates committed code for application-specific vulnerabilities by tracing workflows and testing whether safeguards actually apply. It complements static analysis, dependency scanning, and secret scanning, which the article says cover other useful jobs. A separate verifier checks reachability, protections, and exploitation conditions; findings expose supporting code and uncertainty for developers to assess. The scan begins with a whole-repository baseline, then delta scans revisit changed code and attack-surface points that are stale or unverified. Eligible findings can include runtime reproducer guides; results apply only to the tested environment and conditions. Supported findings can generate remediation pull requests for human review. The article cautions that model verification can share mistaken assumptions, and code alone may not establish external-service or deployment behavior. Coverage statuses record assessment, not a guarantee that the code is vulnerability-free.

### Source excerpt

Attackers now point AI at unfamiliar code to trace workflows and probe the assumptions its authors made. Deep Scan in CodeRabbit Secure does the same for your team first: it investigates committed code, verifies findings independently, and shows the evidence before you decide what to fix.

## CodeRabbit Secure now hunts across repos and beyond the merge

DevFeed: [CodeRabbit Secure now hunts across repos and beyond the merge](<https://devfeed.tech/articles/coderabbit-secure-now-hunts-across-repos-and-beyond-the-merge-103463.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-secure-cross-repository-security>)

Author: Rohit Khanna

Published: 2026-10-27T12:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [code security](<https://devfeed.tech/topics/code-security.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [Static code analysis](<https://devfeed.tech/topics/static-code-analysis.md>)

Tags: [announcements](<https://devfeed.tech/tags/announcements.md>), [api](<https://devfeed.tech/tags/api.md>), [attack-paths](<https://devfeed.tech/tags/attack-paths.md>), [attack-surface](<https://devfeed.tech/tags/attack-surface.md>), [block](<https://devfeed.tech/tags/block.md>), [blocking](<https://devfeed.tech/tags/blocking.md>), [budget](<https://devfeed.tech/tags/budget.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [compliance](<https://devfeed.tech/tags/compliance.md>), [context](<https://devfeed.tech/tags/context.md>), [control](<https://devfeed.tech/tags/control.md>), [cost](<https://devfeed.tech/tags/cost.md>), [coverage](<https://devfeed.tech/tags/coverage.md>), [critical](<https://devfeed.tech/tags/critical.md>), [cross](<https://devfeed.tech/tags/cross.md>), [data-lake](<https://devfeed.tech/tags/data-lake.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [repo](<https://devfeed.tech/tags/repo.md>), [runtime](<https://devfeed.tech/tags/runtime.md>), [scan](<https://devfeed.tech/tags/scan.md>), [security](<https://devfeed.tech/tags/security.md>), [verification](<https://devfeed.tech/tags/verification.md>)

### AI overview

CodeRabbit Secure extends security review from individual pull requests to ongoing, repository-wide coverage. Deep Scan traces risks across linked repositories (up to 20 for context), while delta scans revisit changed code and stale coverage to reduce repeated work and cost. For eligible Deep Scan findings, runtime verification replays exploit attempts in a disposable sandbox; it does not test production systems. On GitHub, admins can require a Security check that blocks merges for open findings at a configured severity until CodeRabbit verifies fixes. Findings can also be routed to Jira or Linear, and Enterprise customers can retrieve Deep Scan results through a read-only API. The excerpt says linked-repository context is available on GitHub Cloud, GitHub Enterprise Server, and GitLab Cloud; merge checks are currently GitHub-only. These features work alongside existing static checks.

### Source excerpt

CodeRabbit Secure now goes beyond the point-in-time scan, with cross-repo security analysis, runtime verification, and enforceable merge checks that keep security connected to your code as it changes.

## For the next two weeks, we're covering $1 million in security scans and fixes

DevFeed: [For the next two weeks, we're covering $1 million in security scans and fixes](<https://devfeed.tech/articles/for-the-next-two-weeks-we-re-covering-1-million-in-security-scans-and-fixes-98980.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/one-million-security-scans-and-fixes>)

Author: Aricka Flowers

Published: 2026-10-24T00:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [code security](<https://devfeed.tech/topics/code-security.md>), [Security](<https://devfeed.tech/topics/security.md>), [Open Source](<https://devfeed.tech/topics/open-source.md>), [open source maintenance](<https://devfeed.tech/topics/open-source-maintenance.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-models](<https://devfeed.tech/tags/ai-models.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [code](<https://devfeed.tech/tags/code.md>), [community](<https://devfeed.tech/tags/community.md>), [cost](<https://devfeed.tech/tags/cost.md>), [covering](<https://devfeed.tech/tags/covering.md>), [deep](<https://devfeed.tech/tags/deep.md>), [draft](<https://devfeed.tech/tags/draft.md>), [english](<https://devfeed.tech/tags/english.md>), [exploitability](<https://devfeed.tech/tags/exploitability.md>), [featured](<https://devfeed.tech/tags/featured.md>), [independent](<https://devfeed.tech/tags/independent.md>), [maintainers](<https://devfeed.tech/tags/maintainers.md>), [members](<https://devfeed.tech/tags/members.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [open-source-projects](<https://devfeed.tech/tags/open-source-projects.md>), [reasoning](<https://devfeed.tech/tags/reasoning.md>), [repositories](<https://devfeed.tech/tags/repositories.md>), [scan](<https://devfeed.tech/tags/scan.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

CodeRabbit is covering up to $5,000 in Deep Scan usage per organization with at least two members through November 8, 2026, while a $1 million allocation lasts on a first-come, first-served basis. Deep Scan examines committed code across a repository, maps code paths and safeguards, and uses semantic analysis plus an independent verifier to assess reachability and exploitability before reporting findings with evidence. It is designed to complement existing security tools, including by analyzing first-party application logic and, where relevant, linked repositories. For eligible findings, it can draft a codebase-aligned fix, but a developer reviews the change and remains responsible for what ships. To use the offer, sign in or create an account, connect a repository, and run a scan.

### Source excerpt

CodeRabbit is making up to $1 million in Deep Scan usage available at no cost until November 8. Any organization with two or more members, from open-source projects to current, former, and new customers, can access up to $5,000 per Organization while the allocation lasts.

## CodeRabbit joins the SpaceX Super Intelligence Partner Network

DevFeed: [CodeRabbit joins the SpaceX Super Intelligence Partner Network](<https://devfeed.tech/articles/coderabbit-joins-the-spacex-super-intelligence-partner-network-96037.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-spacex-super-intelligence-partner-network>)

Author: Blair Pierson

Published: 2026-10-22T00:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [ai-governance](<https://devfeed.tech/topics/ai-governance.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [code security](<https://devfeed.tech/topics/code-security.md>)

Tags: [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-change-management](<https://devfeed.tech/tags/agentic-change-management.md>), [agentic-development](<https://devfeed.tech/tags/agentic-development.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [autonomous](<https://devfeed.tech/tags/autonomous.md>), [build](<https://devfeed.tech/tags/build.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [development](<https://devfeed.tech/tags/development.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [governance](<https://devfeed.tech/tags/governance.md>), [grok](<https://devfeed.tech/tags/grok.md>), [grok-build](<https://devfeed.tech/tags/grok-build.md>), [independent](<https://devfeed.tech/tags/independent.md>), [marketplace](<https://devfeed.tech/tags/marketplace.md>), [partner](<https://devfeed.tech/tags/partner.md>), [plugin](<https://devfeed.tech/tags/plugin.md>), [review](<https://devfeed.tech/tags/review.md>), [spacex](<https://devfeed.tech/tags/spacex.md>), [spacex-super-intelligence](<https://devfeed.tech/tags/spacex-super-intelligence.md>)

### AI overview

CodeRabbit is available in the Grok Build Plugin Marketplace as an independent review and governance layer for changes generated in Grok Build. It reviews work from developers and AI agents against repository context and organizational standards, explains codebase impacts, prioritizes human attention, and investigates vulnerabilities in new and existing code. Teams can review pull requests on the web or use CodeRabbit in an IDE or CLI, then apply their judgment to findings. Dub's five-person team uses Grok Build to generate code and CodeRabbit as a final review layer; its CEO says reviews became 100% to 150% more efficient and describes scoping releases into successive versions as part of the team's risk management. The efficiency figure is his estimate, and the account is a company example.

### Source excerpt

CodeRabbit is now available in the Grok Build Plugin Marketplace, connecting agentic development with independent code review and governance.

## Which software engineering skills matter in the age of AI?

DevFeed: [Which software engineering skills matter in the age of AI?](<https://devfeed.tech/articles/which-software-engineering-skills-matter-in-the-age-of-ai-94468.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/software-engineering-skills-ai-kunal-kushwaha>)

Author: Hendrik Krack

Published: 2026-10-21T00:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Software Engineering](<https://devfeed.tech/topics/software-engineering.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [agentic-coding](<https://devfeed.tech/topics/agentic-coding.md>), [llama.cpp](<https://devfeed.tech/topics/llama-cpp.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [Learning](<https://devfeed.tech/topics/learning.md>), [engineering-culture](<https://devfeed.tech/topics/engineering-culture.md>), [Developer experience](<https://devfeed.tech/topics/developer-experience.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [english](<https://devfeed.tech/tags/english.md>), [fundamentals](<https://devfeed.tech/tags/fundamentals.md>), [implementation](<https://devfeed.tech/tags/implementation.md>), [llms](<https://devfeed.tech/tags/llms.md>), [product-judgment](<https://devfeed.tech/tags/product-judgment.md>), [skills](<https://devfeed.tech/tags/skills.md>), [software-engineering](<https://devfeed.tech/tags/software-engineering.md>)

### AI overview

As AI agents take on more implementation, engineers still need technical fundamentals and judgment to choose problems, evaluate generated code, and remain accountable for what ships. Kunal Kushwaha recommends reading and debugging agent output to deepen understanding, building product judgment by explaining what a website should change and why, and prioritizing a working first version that people can try. He expects agents to handle more code review, though he would still want human review for high-stakes fixes; the discussion also notes that passing tests may not reveal maintenance, performance, security, or wider application impacts. To build technical depth, he suggests running LLMs locally, then exploring fine-tuning and scale to understand infrastructure and deployment tradeoffs among APIs, owned hardware, and managed platforms. Open-source work, hackathons, projects, and sharing work for feedback offer ways to build and demonstrate experience.

### Source excerpt

Kunal Kushwaha on technical fundamentals, product judgment, code review, and running LLMs locally as AI coding agents take on more implementation work.

## CodeRabbit brings independent validation to Cursor Origin

DevFeed: [CodeRabbit brings independent validation to Cursor Origin](<https://devfeed.tech/articles/coderabbit-brings-independent-validation-to-cursor-origin-86227.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-independent-validation-cursor-origin>)

Author: Avery Jennings

Published: 2026-10-13T12:00:00Z

Content type: release

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [cursor origin](<https://devfeed.tech/topics/cursor-origin.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [app](<https://devfeed.tech/tags/app.md>), [assessment](<https://devfeed.tech/tags/assessment.md>), [automated-testing](<https://devfeed.tech/tags/automated-testing.md>), [beta](<https://devfeed.tech/tags/beta.md>), [business-logic](<https://devfeed.tech/tags/business-logic.md>), [change](<https://devfeed.tech/tags/change.md>), [coding-agent](<https://devfeed.tech/tags/coding-agent.md>), [commit](<https://devfeed.tech/tags/commit.md>), [confidence](<https://devfeed.tech/tags/confidence.md>), [connect](<https://devfeed.tech/tags/connect.md>), [cursor](<https://devfeed.tech/tags/cursor.md>), [cursor-origin](<https://devfeed.tech/tags/cursor-origin.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [independent](<https://devfeed.tech/tags/independent.md>), [launch](<https://devfeed.tech/tags/launch.md>), [marketplace](<https://devfeed.tech/tags/marketplace.md>), [partner](<https://devfeed.tech/tags/partner.md>), [pull-requests](<https://devfeed.tech/tags/pull-requests.md>), [teams](<https://devfeed.tech/tags/teams.md>), [validation](<https://devfeed.tech/tags/validation.md>)

### AI overview

CodeRabbit reviews pull requests in repositories hosted natively on Cursor Origin, posting a walkthrough and inline findings for human evaluation. Authors can discuss a finding in its thread, make a correction themselves or ask a coding agent to implement it, and CodeRabbit checks the updated commit; the review conversation stays attached to the PR while the team retains merge authority. Setup requires a workspace admin to connect an Origin namespace, approve repository access, and ensure PR authors have review access. An eligible CodeRabbit Review plan or trial is required. The integration is currently available in CodeRabbit's US region; GitHub repositories synced into Origin should continue using CodeRabbit's GitHub integration. Cursor Origin is gradually rolling out in early beta to users on paid Cursor plans.

### Source excerpt

CodeRabbit is joining Cursor Origin's app marketplace as a launch partner, bringing independent validation to pull requests hosted natively on Origin so teams can ship agent-written code with confidence.

## Open-source reviews on CodeRabbit topped 1 million in a single month

DevFeed: [Open-source reviews on CodeRabbit topped 1 million in a single month](<https://devfeed.tech/articles/open-source-reviews-on-coderabbit-topped-1-million-in-a-single-month-79669.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/one-million-open-source-reviews-september-2026>)

Author: Santosh Yadav

Published: 2026-10-08T12:00:00Z

Content type: news

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Open Source](<https://devfeed.tech/topics/open-source.md>), [open source maintenance](<https://devfeed.tech/topics/open-source-maintenance.md>), [Maintainers](<https://devfeed.tech/topics/maintainers.md>)

Tags: [2026](<https://devfeed.tech/tags/2026.md>), [announcements](<https://devfeed.tech/tags/announcements.md>), [english](<https://devfeed.tech/tags/english.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [pull-request](<https://devfeed.tech/tags/pull-request.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

CodeRabbit reports completing 1,049,817 reviews on open-source repositories in September 2026, a monthly record and 14.4% above August. The article describes its code review tools, support for maintainers, and more than $10 million commitment to open source over 12 months.

### Source excerpt

See which open-source projects ran the most CodeRabbit reviews in September and how we're backing maintainers through a commitment of more than $10 million to open source.

## The feature works. But does it belong?

DevFeed: [The feature works. But does it belong?](<https://devfeed.tech/articles/the-feature-works-but-does-it-belong-79687.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/the-feature-works-does-it-belong>)

Author: Hendrik Krack

Published: 2026-10-08T00:00:00Z

Content type: news

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [AI-enhanced product development](<https://devfeed.tech/topics/ai-enhanced-product-development.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Product Management](<https://devfeed.tech/topics/product-management.md>), [Test automation](<https://devfeed.tech/topics/test-automation.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [feature bloat](<https://devfeed.tech/topics/feature-bloat.md>), [User interface design](<https://devfeed.tech/topics/ui-design.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [about](<https://devfeed.tech/tags/about.md>), [accelerated](<https://devfeed.tech/tags/accelerated.md>), [added](<https://devfeed.tech/tags/added.md>), [advice](<https://devfeed.tech/tags/advice.md>), [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-assisted](<https://devfeed.tech/tags/ai-assisted.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [alerts](<https://devfeed.tech/tags/alerts.md>), [allowed](<https://devfeed.tech/tags/allowed.md>), [but](<https://devfeed.tech/tags/but.md>), [development](<https://devfeed.tech/tags/development.md>), [english](<https://devfeed.tech/tags/english.md>), [feature](<https://devfeed.tech/tags/feature.md>), [labs](<https://devfeed.tech/tags/labs.md>), [shipping](<https://devfeed.tech/tags/shipping.md>), [take](<https://devfeed.tech/tags/take.md>), [trm-labs](<https://devfeed.tech/tags/trm-labs.md>), [validating](<https://devfeed.tech/tags/validating.md>), [work](<https://devfeed.tech/tags/work.md>), [works](<https://devfeed.tech/tags/works.md>)

### AI overview

Both founders describe agents making implementation faster, which raises the stakes of deciding what to build and how to review it. TRM Labs uses a "product cortex" to turn customer feedback into prioritized ideas, requirements, and prototypes; adding its design system and strategic priorities improved agent suggestions, and Rahul estimated about 10% of signals became work the team would eventually build. TRM required human approval for customer-facing pull requests and deployments while developing automated test evidence and controls to route security-sensitive changes for human review. CodeRabbit is exploring an approval step before background agents open pull requests and senior review of prompts; developers remain responsible for validating changes. Faster output also pushed CodeRabbit to improve build and delivery infrastructure. Both teams let product and design contribute directly to code while engineers retain responsibility for decisions such as scale and API design.

### Source excerpt

TRM Labs' Rahul Raina and CodeRabbit's Harjot Gill on choosing, validating, and shipping work as agents take on more development.

## Introducing collaborative coding: Build together with Slack Code and CodeRabbit Agent for Slack

DevFeed: [Introducing collaborative coding: Build together with Slack Code and CodeRabbit Agent for Slack](<https://devfeed.tech/articles/introducing-collaborative-coding-build-together-with-slack-code-and-coderabbit-agent-for-slack-79661.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/introducing-collaborative-coding>)

Author: Arvind Soni

Published: 2026-10-07T00:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Slack](<https://devfeed.tech/topics/slack.md>), [Spec Driven Development](<https://devfeed.tech/topics/spec-driven-development.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [CI/CD](<https://devfeed.tech/topics/cicd.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agents](<https://devfeed.tech/tags/agents.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-coding](<https://devfeed.tech/tags/ai-coding.md>), [ai-coding-agents](<https://devfeed.tech/tags/ai-coding-agents.md>), [align](<https://devfeed.tech/tags/align.md>), [analysis](<https://devfeed.tech/tags/analysis.md>), [api](<https://devfeed.tech/tags/api.md>), [artifacts](<https://devfeed.tech/tags/artifacts.md>), [beta](<https://devfeed.tech/tags/beta.md>), [branch](<https://devfeed.tech/tags/branch.md>), [build](<https://devfeed.tech/tags/build.md>), [code](<https://devfeed.tech/tags/code.md>), [coding](<https://devfeed.tech/tags/coding.md>), [cron](<https://devfeed.tech/tags/cron.md>), [database](<https://devfeed.tech/tags/database.md>), [database-schema](<https://devfeed.tech/tags/database-schema.md>), [english](<https://devfeed.tech/tags/english.md>), [slack](<https://devfeed.tech/tags/slack.md>), [slack-code](<https://devfeed.tech/tags/slack-code.md>), [spec](<https://devfeed.tech/tags/spec.md>)

### AI overview

Collaborative coding puts relevant experts and AI agents in one Slack Code channel so specifications, implementation, and review share the same context. In the example, a Salesforce admin defines authorized fields and security, a product engineer supplies database knowledge, and an infrastructure engineer joins for deployment; each can catch up from the channel history or a CodeRabbit summary. Slack Code provides artifacts, a canvas, and a code diff view. CodeRabbit Agent for Slack can investigate Jira or Linear requests, refine specs against a codebase, identify code experts, and use cloud coding agents (in beta) to implement, commit, and open pull requests. It reviews opened pull requests and can prepare or apply changes from review comments; a human approver and the repository's protected merge process remain part of finalization. The described trade-offs are adapting to a new collaboration habit, using cloud agents instead of local development, and an evolving Slack Code workflow. Getting started requires a CodeRabbit trial, a connected active Git repository, and the Slack agent installation; existing customers can skip the first two steps.

### Source excerpt

Bring the people who know each part of a coding task into one shared Slack Code channel, where they can shape the spec and build with CodeRabbit Agent for Slack.

## How CodeRabbit Security performed on a real-world vulnerability benchmark

DevFeed: [How CodeRabbit Security performed on a real-world vulnerability benchmark](<https://devfeed.tech/articles/how-coderabbit-security-performed-on-a-real-world-vulnerability-benchmark-79677.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/security-agent-benchmarks>)

Author: Rohit Khanna

Published: 2026-10-06T12:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [Benchmark](<https://devfeed.tech/topics/benchmark.md>), [Exploit](<https://devfeed.tech/topics/exploit.md>), [Security](<https://devfeed.tech/topics/security.md>), [gaia](<https://devfeed.tech/topics/gaia.md>)

Tags: [ai-agents](<https://devfeed.tech/tags/ai-agents.md>), [benchmark](<https://devfeed.tech/tags/benchmark.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [english](<https://devfeed.tech/tags/english.md>), [evaluation](<https://devfeed.tech/tags/evaluation.md>), [real-world](<https://devfeed.tech/tags/real-world.md>), [security](<https://devfeed.tech/tags/security.md>), [vulnerability](<https://devfeed.tech/tags/vulnerability.md>)

### AI overview

CodeRabbit compares its Security product with three other systems on recovering known vulnerabilities in open-source repositories. Its 100-case benchmark reports detection rates of 86.7% for CodeRabbit, 73.3% for Devin, 62.0% for Codex Security Plugin, and 40.0% for Claude Security. The article describes its test setup, scoring method, and staged map, hunt, and verify workflow, while noting that fix quality is outside the reported score and prior exposure to public vulnerabilities remains a limitation.

### Source excerpt

CodeRabbit Security led three other systems in our vulnerability-detection benchmark. Here's what we tested and how we scored it.

## How independent AI code review builds trust in agent-generated changes

DevFeed: [How independent AI code review builds trust in agent-generated changes](<https://devfeed.tech/articles/how-independent-ai-code-review-builds-trust-in-agent-generated-changes-79698.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/why-agentic-change-management-starts-with-independent-ai-code-review>)

Author: Brandon Gubitosa

Published: 2026-10-05T07:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [human review](<https://devfeed.tech/topics/human-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>)

Tags: [agent](<https://devfeed.tech/tags/agent.md>), [agentic](<https://devfeed.tech/tags/agentic.md>), [agentic-change-management](<https://devfeed.tech/tags/agentic-change-management.md>), [agentic-workflows](<https://devfeed.tech/tags/agentic-workflows.md>), [ai](<https://devfeed.tech/tags/ai.md>), [ai-code](<https://devfeed.tech/tags/ai-code.md>), [ai-code-review](<https://devfeed.tech/tags/ai-code-review.md>), [api](<https://devfeed.tech/tags/api.md>), [authorization](<https://devfeed.tech/tags/authorization.md>), [capabilities](<https://devfeed.tech/tags/capabilities.md>), [changes](<https://devfeed.tech/tags/changes.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [control](<https://devfeed.tech/tags/control.md>), [english](<https://devfeed.tech/tags/english.md>), [governance](<https://devfeed.tech/tags/governance.md>), [independent](<https://devfeed.tech/tags/independent.md>), [layer](<https://devfeed.tech/tags/layer.md>), [review](<https://devfeed.tech/tags/review.md>), [teams](<https://devfeed.tech/tags/teams.md>), [trust](<https://devfeed.tech/tags/trust.md>)

### AI overview

Independent AI review gives teams a separate source of evidence for deciding whether agent-generated changes are ready to ship. CodeRabbit clones each repository into an isolated sandbox, builds a code graph, runs applicable tools from a catalog of more than 50 linters and security integrations, and adds context such as team guidelines, code indexes, issue history, CI/CD state, and relevant web information. It uses multiple models for different review tasks and checks findings against available evidence before posting them. This can expose assumptions that passing tests miss, while leaving human reviewers to assess the findings, investigate unresolved risks, and make the merge decision. CodeRabbit presents Review alongside Triage, Explain, and Secure as parts of its governance and control layer; the excerpt does not detail those other capabilities.

### Source excerpt

Independent AI code review gives teams evidence to trust proposed changes, supported by CodeRabbit's governance and control layer for agentic software development.

## The Merge: What if your 'if-statements' could think?

DevFeed: [The Merge: What if your 'if-statements' could think?](<https://devfeed.tech/articles/the-merge-what-if-your-if-statements-could-think-79691.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/the-merge-typesafe-jev-smart-if-statements>)

Author: Hendrik Krack

Published: 2026-10-01T00:00:00Z

Content type: article

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [AI interaction best practices](<https://devfeed.tech/topics/ai-interaction-best-practices.md>), [if statement](<https://devfeed.tech/topics/if-statement.md>), [AI-assisted coding](<https://devfeed.tech/topics/ai-assisted-coding.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>), [AI for customer service automation](<https://devfeed.tech/topics/ai-for-customer-service-automation.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [api](<https://devfeed.tech/tags/api.md>), [aws-lambda](<https://devfeed.tech/tags/aws-lambda.md>), [batching](<https://devfeed.tech/tags/batching.md>), [classification](<https://devfeed.tech/tags/classification.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [decision-models](<https://devfeed.tech/tags/decision-models.md>), [docker](<https://devfeed.tech/tags/docker.md>), [drop-in](<https://devfeed.tech/tags/drop-in.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [if](<https://devfeed.tech/tags/if.md>), [jev](<https://devfeed.tech/tags/jev.md>), [typesafe](<https://devfeed.tech/tags/typesafe.md>)

### AI overview

Jev is TypeSafe's decision model for classification, scoring, and routing: it takes application state plus typed questions and returns a defined choice, rubric score, or yes/no probability, avoiding generated text that software must parse. Allie Laabs says it can fit existing LLM-based classification tasks and branching decisions that are difficult to specify mathematically; it is not a replacement for general language models. Jev answers questions independently in parallel, so asking potentially relevant follow-up questions in one request can avoid resending state. A TypeSafe cookbook reports a 13-question batch was 12.2 times cheaper and 10 times faster than separate calls, with unchanged answers. Laabs says coding agents often need a review rule to bundle calls that send the same state. TypeSafe reports $0.042 per million input tokens and 70-500 ms response times, while noting its latency tests were near West Coast servers. Its known-limitations page warns that injected instructions or persuasive text in the state can shift classifications, so low-confidence answers need fallback paths and action-authorizing decisions require care.

### Source excerpt

TypeSafe's Allie Laabs explains how to find Jev-shaped problems in your code, why you should ask every question up front, and why coding agents still need help getting it right.

## CodeRabbit joins the OpenAI Marketplace

DevFeed: [CodeRabbit joins the OpenAI Marketplace](<https://devfeed.tech/articles/coderabbit-joins-the-openai-marketplace-79625.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/coderabbit-openai-marketplace>)

Author: Blair Pierson

Published: 2026-09-29T00:00:00Z

Content type: news

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [aws-marketplace](<https://devfeed.tech/topics/aws-marketplace.md>), [engineering-leadership](<https://devfeed.tech/topics/engineering-leadership.md>)

Tags: [announcements](<https://devfeed.tech/tags/announcements.md>), [coding-agents](<https://devfeed.tech/tags/coding-agents.md>), [english](<https://devfeed.tech/tags/english.md>), [marketplace](<https://devfeed.tech/tags/marketplace.md>), [openai](<https://devfeed.tech/tags/openai.md>), [review](<https://devfeed.tech/tags/review.md>), [security](<https://devfeed.tech/tags/security.md>)

### AI overview

OpenAI customers can apply existing OpenAI spend commitments to purchase CodeRabbit through the OpenAI Marketplace. CodeRabbit reviews code from developers and AI coding agents, explains changes across a codebase, prioritizes human review, and investigates security vulnerabilities. People decide what to approve and release.

### Source excerpt

OpenAI customers can apply their existing OpenAI spend commitment towards CodeRabbit through the OpenAI Marketplace, simplifying procurement.

## Claude Sonnet 5.5 for code review: More catches than Sonnet 5, in half the time

DevFeed: [Claude Sonnet 5.5 for code review: More catches than Sonnet 5, in half the time](<https://devfeed.tech/articles/claude-sonnet-5-5-for-code-review-more-catches-than-sonnet-5-in-half-the-time-79680.md>)

Original publisher: [Read original article](<https://coderabbit.ai/blog/sonnet-5-5-model-review>)

Author: Hendrik Krack

Published: 2026-09-28T00:00:00Z

Content type: news

Language: en

Sources: [CodeRabbit](<https://devfeed.tech/sources/coderabbit.md>)

Topics: [ai code review](<https://devfeed.tech/topics/ai-code-review.md>), [Code review](<https://devfeed.tech/topics/code-review.md>), [Pull Request](<https://devfeed.tech/topics/pull-request.md>)

Tags: [ai](<https://devfeed.tech/tags/ai.md>), [bugs](<https://devfeed.tech/tags/bugs.md>), [claude-sonnet-5-5](<https://devfeed.tech/tags/claude-sonnet-5-5.md>), [code](<https://devfeed.tech/tags/code.md>), [code-review](<https://devfeed.tech/tags/code-review.md>), [costs](<https://devfeed.tech/tags/costs.md>), [english](<https://devfeed.tech/tags/english.md>), [featured](<https://devfeed.tech/tags/featured.md>), [opus-5-5](<https://devfeed.tech/tags/opus-5-5.md>), [review](<https://devfeed.tech/tags/review.md>), [sonnet-5](<https://devfeed.tech/tags/sonnet-5.md>), [time](<https://devfeed.tech/tags/time.md>)

### AI overview

In CodeRabbit's 13 hard known-bug cases, Claude Sonnet 5.5 caught 6 issues versus Sonnet 5's 4, at similar actionable precision (41.2% vs. 40.0%), but the models missed different bugs and the sample is small. Across 44 open-source pull requests, Sonnet 5.5 averaged 6:33 per review versus 13:31 and posted 24% fewer comments; judged coverage for that larger set was still pending. Its Claude calls cost about 40% as much per review as Sonnet 5's at list prices. Sonnet 5.5 remains behind Opus 5.5 on the hard cases (6 catches versus 8-10). Adaptive thinking caught one more case than thinking off in the small comparison, with slightly higher precision, about twice the output tokens, and no observed slowdown. The authors recommend evaluating it on pull requests with known outcomes and measuring the full review pipeline, including verification and summary models.

### Source excerpt

We benchmarked Claude Sonnet 5.5 for AI code review against Sonnet 5 and Opus 5.5: which bugs it catches, how fast it reviews, and what a review costs.

[Next page](<https://devfeed.tech/sources/coderabbit.md?cursor=WyIyMDI2LTA5LTI4VDAwOjAwOjAwKzAwOjAwIiwgIjk0NjAzYTk1LWI0NmItNGFkNi05NWEyLWFhOGZkN2ZiN2MyMCJd>)