In short:
ak-code-reviewpicks exactly what needs reviewing, checks it against the plan or spec, then reviews quality, and returns a list of findings with file, line, impact and evidence. When it is done you move on toak-ship, or toak-fixafter a whole-codebase audit. A review-only run does not edit code, post to the PR or merge.
Disclosure: I maintain AgentKit and have an affiliate relationship with it: the AgentKit link in this post is an affiliate link; buying through it gets you 30% off.
Summary
Use it when you have just written code and want to know whether it has bugs, breaks something else, or misses the requirements: before a commit, before merging a PR, after a high-risk commit, before a big release, or when you need to check someone else's review comments. You do not need the full workflow for a tiny change like a typo fix, and do not expect it to fix things unless you allow it.
flowchart LR
A["Code written, tests run"] --> B["ak-code-review"]
B --> C["Findings with file, line, evidence"]
C --> D["ak-ship"]
C --> E["You allow fixes, then it reviews again"]
C --> F["Stop, report only"]It revolves around 4 points:
| Point | Question | Example |
|---|---|---|
| Input | What is being reviewed this time? | #<PR number> or a PR link, a commit hash of 7 or more hex characters, --pending for uncommitted changes, codebase to scan the whole repo, codebase parallel to have several reviewers check together. With nothing given, it looks at the changes just made in the session, and if there are none, it asks you. |
| Right first, then good | Does the code do what the plan or spec asks? | The plan lists 5 requirements and the code has 4: the missing one is marked MISSING and must be added before the quality review. Code the spec never mentions is marked EXTRA and proposed for removal, unless there is a reason to keep it. |
| Findings with evidence | Where is the bug, what breaks, and what backs that claim? | [src/api/search.ts:18] the search term is concatenated into the SQL query, so anyone can read other tables. A fix comes with it. A score never replaces a check command or a concrete finding. |
| Review only means review only | Does it edit code, comment on the PR or merge by itself? | No. You have to say clearly that it may fix. Then it fixes only confirmed findings, within scope, and reviews the fixed version again. |
The scope you asked for is treated as a constraint, not a defect to cut: it only proposes trimming with --yagni. It also never claims code is correct in places it has not looked at.
Sample prompt before a commit:
/ak-code-review --pending
Purpose: fix how the order total is calculated when a discount code is applied in <app name>.
Constraints: do not change the price format returned to the old API.| Flag | What it is for | Since ak |
|---|---|---|
--pending | Reviews uncommitted changes | 1.0.0 |
--ultra | 5 independent reviewers, keeps every verified bug | 2.13.0 |
--advice | A stronger model reviews at key checkpoints | 2.13.0 |
--yagni | Allows proposals to cut scope | 2.8.0 |
Below is the detailed version: the run flow, 5 situations with prompts, each flag explained (when to use it, when you do not need it, what you get, which flags it combines with), the change history by ak release, and common pitfalls. The detailed version and the interactive page are for subscribers. Your agent can also search them through the portal's MCP.
Opens when your account has subscriber access
This guide is for subscribers
Once your account has access, the full piece, the interactive visual and the agent call all open.
- Sign in or create a modernEngineer accountReading access belongs to your account email.
- Choose a plan on the pricing page1, 3 or 12 months, paid by VietQR bank transfer. Access opens about a minute after the money arrives.
- Or redeem a Facebook subscriber codePeople supporting the page on Facebook get an activation code by message each month.
Already a subscriber? Sign in to read.
Use this piece in your agent
Get an API keyCopy the text below and paste it into Claude Code, Codex or whichever agent you use. The agent reads the piece and checks it against the repo you have open.
Read "ak-code-review: find bugs and regressions in code, with evidence" on modernEngineer and apply it to this repo.
1. Call the content_get tool of the modernengineer MCP server with locale "en", slug "ak-code-review". If the modernengineer MCP server is not connected, stop and walk me through connecting it from https://modernengineer.dev/en/mcp.
2. Summarize the main steps of the piece and point out which apply to this repo.
3. Propose concrete changes and ask me before editing any file.Connect MCP in Claude Code (once)
claude mcp add --transport http modernengineer \
https://api.modernengineer.dev/mcp \
--header "Authorization: Bearer <API key>"Reading the full piece over MCP needs an API key from a subscriber account or one that bought this guide. Without access, the agent only gets the opening.
- Version
- v1.0.0
- Updated
- 22/10/2026
- Licence
- All rights reserved
- Slug
- ak-code-review
