Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
Muse
Dots+21 moreClaude CodeCursorCodex CLIMuseDots
Manus
Grok
Grok Bot+18 agentsSwitch ai Code Reviewer
Audit local git diffs for memory leaks, async bottlenecks, and exposed security keys before pushing code.
$6
switch-ai-code-reviewer
Example session with this skill installed
Run triage on my latest git diff for the auth-service.js file. I am worried about the new async login flow and potential token leakage.
- Read your context and instructions
- Compiled the switch-ai-code-reviewer
I have completed the triage and security audit for the auth-service.js diff.
PR-Triage & Security Audit Report
| Category | Priority | Issue | Location | Recommendation |
|---|---|---|---|---|
| Security | CRITICAL | Hardcoded JWT Secret | line 42 | Remove string literal; use process.env.JWT_SECRET. |
| Performance | HIGH | Unawaited Async Call | line 89 | db.logs.save() is called without await, risking a race condition. |
| Memory | MEDIUM | Growing Closure Scope | line 112 | Event listener inside a loop without cleanup. Move handler outside loop. |
| Triage | INFO | Logic Change | line 55 | Updated password hashing rounds from 10 to 12. |
Suggested Fix for Async Bottleneck
// Before
async function login(user) {
const token = generateToken(user);
db.logs.save({ action: 'login', userId: user.id }); // Missing await
return token;
}
// After
async function login(user) {
const token = generateToken(user);
await db.logs.save({ action: 'login', userId: user.id });
return token;
}
Next steps
- Move the hardcoded secret at line 42 to your
.envfile. - Apply the
awaitfix to the logging call to ensure data consistency. - Re-run the audit tool once changes are saved to verify the fix.
Connects securely to your tools. The creator never sees your data.
About this skill
Staging messy pull requests without a second set of eyes leads to production bugs and leaked secrets. This skill audits local git diffs to catch memory leaks, async bottlenecks, and exposed credentials before you commit.
What it does
- Security Auditing scans code for hardcoded API keys, secrets, and common vulnerability patterns.
- Performance Analysis identifies inefficient async/await patterns and potential memory leaks in logic.
- Diff Triage summarizes changes and flags high-risk modifications for manual oversight.
- Automated Refactoring suggests concrete code improvements to resolve identified bottlenecks.
How it works
- Use the triage command to analyze the current staged or unstaged git diff.
- Review the generated audit report highlighting security risks and performance issues.
- Apply suggested fixes for async patterns or memory management directly to the file.
- Run a final audit on specific files to ensure all vulnerabilities are cleared before pushing.
Frameworks & tools
This tool interfaces with local git environments and uses an MCP server running on Node.js to process code analytics.
Why this beats prompting it yourself
Generic prompts often miss context-specific async bottlenecks or fail to scan for obscure secret patterns. This skill provides a structured audit framework designed specifically for local diff analysis and PR preparation.
Use cases
- Auditing a feature branch for exposed environment variables before a merge.
- Identifying race conditions in complex asynchronous JavaScript or TypeScript logic.
- Generating a summary of changes to assist in writing detailed pull request descriptions.
- Reducing technical debt by catching memory leaks during the development cycle.
Known limitations
Requires a local git repository and Node.js environment to run the MCP server. Performance scanning is optimized for standard async patterns and may not catch every edge case in niche libraries.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Code Quality & Review.
Requires local git and Node.js. Optimized for async/memory audits; might not catch logic-specific bugs in non-standard libraries.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed 4 days ago
- Accounts
GitHub
Switch Omni
- Software
Node.js (v18.0.0 or higher)
Git (initialized local repository)
Permissions
Allowed Hosts
File Scopes
This skill is a complete MCP port of the switch-ai-code-reviewer bot. It has been refactored from a centralized webhook architecture into a secure, client-side stdio MCP server designed for seamless integration with Claude Code and Cursor IDE.