Backend Code Review
Helps with backend development, code review as part of developing backend services and APIs workflows.
Install on your platform
We auto-selected Claude Code based on this skill’s supported platforms.
Run in terminal (recommended)
claude mcp add backend-code-review npx -- -y @trustedskills/backend-code-review
Or manually add to ~/.claude/settings.json
{
"mcpServers": {
"backend-code-review": {
"command": "npx",
"args": [
"-y",
"@trustedskills/backend-code-review"
]
}
}
}Requires Claude Code (claude CLI). Run claude --version to verify your install.
About This Skill
What it does
This skill assists with reviewing, analyzing, and improving backend code written in Python (.py) located within the api/ directory. It supports three review modes: pending changes (reviewing files slated for commit), code snippets provided by the user, and focused reviews of specific files. The goal is to provide actionable fixes and suggestions based on defined checklists or general review rules.
When to use it
- Reviewing current changes before committing them.
- Getting feedback on a function, class, or module excerpt you've written.
- Analyzing specific backend files for potential improvements (e.g.,
api/app.py).
Key capabilities
- Pending-change review: Inspects staged/working-tree files.
- Code snippets review: Analyzes code provided directly in the chat.
- File-focused review: Reviews specific files or sets of files within the
api/directory. - Checklist-driven reviews: Follows rules for database schema design, architecture, repository abstraction, and SQLAlchemy patterns.
- General Review Rules: Includes security (SQL injection, SSRF, etc.) and performance checks (N+1 queries, missing indexes).
Example prompts
- "Can you review the changes I've staged for commit?"
- "Please review this function:
def calculate_something(): ..." - "Could you take a look at
api/users.pyand let me know if there are any improvements to be made?"
Tips & gotchas
- The skill is specifically designed for backend code within the
api/directory. It does not handle frontend code (e.g., .tsx, .js). - Always provide specific files or code snippets for review; avoid broad requests.
- Actionable fixes and suggestions are expected in the output, ideally with file and line number references when available.
Tags
TrustedSkills Verification
Unlike other registries that point to live repositories, TrustedSkills pins every skill to a verified commit hash. This protects you from malicious updates — what you install today is exactly what was reviewed and verified.
Security Audits
| Gen Agent Trust Hub | Pass |
| Socket | Pass |
| Snyk | Pass |
🌐 Community
Passed automated security scans.