Skip to content
Scaffolding
Skill

/review-changes

Review changes against file-specific toolkit rules. Trigger for code-review and quality-check requests. Do not trigger when the user asks to generate a project or author coding rules.

BOOST
From plugin
aicode-toolkit
1619 skills1 command6 MCP
Install
$ npx -y skills add agiflow/aicode-toolkit --skill review-changes --agent claude-code

How it fires

How this skill gets triggered: by you, by Claude, or both.

  • Fires itselfAuto-invocation. Claude auto-loads it when your prompt matches the work.Auto-invocation is when the right skill fires by itself at the right moment, driven by a FLOW.md router and a hook, instead of you invoking it by name. It is the difference between a skill being installed and a skill actually getting used.Read the full definition →
  • You can call itInvoke it directly when you want it.
  • Slash command/review-changes

Context preview

The summary Claude sees to decide when to auto-load this skill.

Review changes against file-specific toolkit rules. Trigger for code-review and quality-check requests. Do not trigger when the user asks to generate a project or author coding rules.

SKILL.md

review-changes.SKILL.md
name: review-changes
description: 'Review changes against file-specific toolkit rules. Trigger for code-review and quality-check requests. Do not trigger when the user asks to generate a project or author coding rules.'

Review changes using project rules

1. Identify changed files and the consumer workspace. For each file call `review-code-change` with `{ "file_path": "..." }` and inspect its full feedback and returned `rules`. 2. Compare actual source and diff against the applicable rules yourself. An unconfigured backend returns rules for **host-agent review** with `identified_issues: []` and `fix_required: false`; those fields are not evidence of compliance. Missing project, template or RULES.yaml likewise means review could not establish compliance. 3. Report findings with file/line, severity, specific rule, evidence and suggested correction. Explicitly state when rules were missing or a backend could not run. Do not claim that an empty issue list passed review without examining the code. 4. Propose fixes and ask permission before editing; recheck any corrected files. Workspace settings may invoke an external agent backend and incur credentials/cost even without CLI backend flags. Check and explain configured settings before requesting such a review.

Ships withaicode-toolkit

This repo provides: project and feature scaffolding via templates file-level design guidance before edits rule-based review after edits design-system discovery for frontend work

Get the whole plugin

Other skills on aicode-toolkit.