> For the complete documentation index, see [llms.txt](https://help.aikido.dev/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://help.aikido.dev/changelog/q3-2026/deep-review-the-agentic-pr-review-that-cuts-merge-time-and-bugs-shipped.md). # Deep Review: the agentic PR review that cuts merge time and bugs shipped Deep Review is Aikido's agentic reviewer for pull requests. It reasons across your full codebase and the discussion already happening on the PR, then posts findings and suggested fixes directly on the pull request, the way a senior security engineer would if they had time to read every change. {% embed url="" %} #### What it does * **Reviews with full context.** Sees what you've discussed or flagged on the PR, plus what your other checks already flagged, not just the diff. * **Reasons across repos.** It traces logic across files, including middleware and global config, and pulls in the linked repositories a change depends on. * **Posts fixes, not just comments.** Findings land on the PR with a suggested fix attached. * **Saves feedback to memory.** Mark a finding as an accepted risk or a false positive, and Deep Review will remember that and learn from it. * **Closes its own findings.** When you push a fix, Deep Review confirms it actually resolves the issue, then resolves the comment for you. #### What it catches Deep Review flags issues that come from how a change interacts with the rest of your app, the type that usually surfaces in a pentest or in production: * **Deeply nested security issues** * **Broken access control** * **Business logic flaws** * **Cross-tenant data leaks** Security is just part of it. Deep Review also flags bugs, unintended side effects, and suggests concrete improvements on every pull request. #### How it fits into Aikido * Runs after your other PR checks (like SAST and dependency checks) so it can build on their findings instead of repeating them. * Shifts deep security review left in your development workflow, catching before merge what a pentest would only find later. It’s the continuous complement to your one-off Code Security Audit. Deep Review is enabled per repository. [Link related repos](https://help.aikido.dev/deep-review/link-repositories) so it reasons across your whole codebase, and [set the review level](https://help.aikido.dev/deep-review/configure-deep-review) that fits your team. #### How much does it cost? Deep Review runs on credits from your Aikido Wallet: 1 credit ($1) per PR, with unlimited re-reviews on the same PR. This makes it 5-10x more accessible than the same review done in-house or with competitors. Get an estimate for how much [it'll cost across your repos](https://app.aikido.dev/repositories/deep-review), and [set a monthly cap](https://app.aikido.dev/settings/wallet/settings) to keep spend predictable. → Read more about [Deep Review](https://help.aikido.dev/deep-review/configure-deep-review) --- # Agent Instructions This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com. ## Querying This Documentation If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question. Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter: ``` GET https://help.aikido.dev/changelog/q3-2026/deep-review-the-agentic-pr-review-that-cuts-merge-time-and-bugs-shipped.md?ask=&goal= ``` `ask` is the immediate question: it should be specific, self-contained, and written in natural language. `goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal. The response will contain a direct answer to the question and relevant excerpts and sources from the documentation. Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.