review-changes

review-changes is a command for Claude Code from unclutter-pro/atlas. It costs 12 tokens per session (671 once invoked), scanned A, original, MIT.

A command that reviews the code changes currently staged or present on a branch, using separate checks for areas such as security, architecture, quality, performance, and documentation.

In plain words
What is it for?
Use it before merging a branch or pull request to find direct issues and problems that changes may cause elsewhere in the project.
Why use it?
It brings together focused reviews and produces one summary of the most noteworthy problems, including affected files.

Command for Claude Code

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods
npx agentmods add commands/unclutter-pro/atlas/review-changes
Clone the repo
git clone --depth 1 https://github.com/unclutter-pro/atlas

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

agentmods badge for review-changes

README.md
[![agentmods](https://agentmods.dev/badge/commands/unclutter-pro/atlas/review-changes.svg)](https://agentmods.dev/commands/unclutter-pro/atlas/review-changes)
Your own site
<a href="https://agentmods.dev/commands/unclutter-pro/atlas/review-changes"><img src="https://agentmods.dev/badge/commands/unclutter-pro/atlas/review-changes.svg" alt="Measured on agentmods" height="20"></a>
Per session 12 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 671 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

ModelPer sessionOnce invoked
Fable 5 $0.00012 $0.00671
Opus 5 $0.00006 $0.00336
Sonnet 5 $0.00002 $0.00134
Haiku 4.5 $0.00001 $0.00067

Measured 3d ago against content hash c1d16ee4ae3e, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

review-changes scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 3d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

.claude/commands/review-changes.md · 86 lines

What it actually says

Review Process

Perform a comprehensive code review using subagents for key areas:

  • architecture-reviewer
  • code-quality-reviewer
  • performance-reviewer
  • documentation-reviewer
  • security-code-reviewer

Instruct each to only provide noteworthy feedback. Take cross-impacted code into consideration. Once they finish, review the feedback and create a feedback summary for the most noteworthy feedback. Do reference files clearly.

# [Short feedback title]

[Short summary with most noteworthy feedback]

## Direct Issues

[The following is listing each issue which are directly related to the changes. In case no issue found just say "Approval, no issue found."]

### 1. [Short title explaining the type of issue]

[Short but detailed description of the effects on the misbehaving / wrong / impacted code with clear reference what files are impacted. No need to tell about prioities or difficulity of the issue. Creator should be able to easily understand the issue.]

### 2. [Next issue ...]

[...]

## Cross Impact Issues

[A list or summary of issues caused by the changes in other parts of the code base. Ignore this section when non-found.]

### [Optional Section:] General Observations

[Additional general observations, not tied to a specific line of code. May include architectural concerns, unclear requirements that need discussion]

### [Optional Section:] Missing Details Issues

[List of files missed on related changes (e.g. missing documentation) or missing test caseses]

## References

[List to important file references, web research references or references to prepared AI prompts for agents.]

In case there is any existing issues/violations, please create a comprehensive temporary markdown document in ./tmp folder (within project). The document should contain a prompt for AI agents, to easily fix these issues. If multiple big issues existing, may split it into multiple files.

Please use the following format for the AI prompt preparations:

Check if this issue is valid — if so, understand the root cause and fix it.

At repo/path/to/file.ts there is an issue ...

[Note: May multiple files can be referenced at once, to give full details. Goal is that AI can fully understand the issue and help fix it. Its formated as text, so copy-paste can be done easily.]

<file name="repo/path/to/file.ts">

<violation number="1" location="repo/path/to/file.ts:86">
[Here comes a more close explaination what violation/issue actually exist. It can exist more then one violation related to the issue. May reference code shortly and explain what cross impact it can have.]
</violation>

<violation number="2" location="repo/path/to/file.ts:210">
[Other violations in the same file, if applicable.]
</violation>

</file>

Keep feedback very concise and precision crafted for the actual issues. Stick to the formats above.


Git Status:

!git status

Please clearly focus on the changes in $ARGUMENTS

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

  1. 3d ago First seen · 86 lines · 12 tokens per session scan A c1d16ee4ae3e

Subscribe to this mod's changes

review-changes is a command published in the GitHub repository unclutter-pro/atlas (2 stars, last pushed 15d ago), licensed MIT. It adds 12 tokens to every session and 671 once invoked, about $0.0001 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.