- Home
- Skills
- Frontend & Web Apps
- mcp devtools
mcp devtools
Audit frontend accessibility, analyze JS bundle sizes, document React props, and run Lighthouse performance reports.
Free
Works with the AI Agents you already use
mcp devtools
Example session with this skill installed
Analyze my React bundle using the stats.json file in the build directory. Check for duplicate packages and large modules.
- Read your context and instructions
- Compiled the mcp devtools
I have analyzed the stats.json file in your build directory. The analysis identified two critical duplicate packages and three modules contributing to over 40% of the total bundle size.
Bundle Analysis Report
Duplicate Packages
| Package | Versions Found | Estimated Impact | Suggested Action |
|---|---|---|---|
lodash | 4.17.15, 4.17.21 | 72KB | Run npm dedupe or update root dependencies. |
date-fns | 2.28.0, 2.29.3 | 45KB | Align versions in package.json. |
Top Bundle Contributors
| Module Path | Size (Gzipped) | % of Total | Recommendation |
|---|---|---|---|
./node_modules/chart.js | 154KB | 18.2% | Use dynamic imports (React.lazy). |
./node_modules/moment | 98KB | 11.5% | Replace with dayjs or date-fns. |
./src/components/HeavyDataGrid | 85KB | 10.1% | Code-split at the route level. |
Next steps
- Run
npm dedupeto resolve the minor version mismatch inlodash. - Implement dynamic imports for the
HeavyDataGridcomponent to reduce initial load time. - Review
package.jsonto lockdate-fnsto a single version across all workspaces.
Connects securely to your tools. The creator never sees your data.
About this skill
Frontend performance and quality issues often remain hidden until they impact users. This skill provides a focused toolkit to audit accessibility, analyze JavaScript bundle bloat, document React component APIs, and run Lighthouse performance reports.
What it does
- Audit accessibility identifies WCAG 2.1 AA violations across entire pages or specific selectors.
- Analyze bundle size identifies large modules and duplicate packages causing bundle bloat.
- Document components scans React directories to generate prop tables with types and requirements.
- Run Lighthouse reports measures Core Web Vitals, SEO, and performance opportunities.
How it works
- Identify the goal by selecting the specific tool needed for accessibility, bundles, documentation, or performance.
- Execute the audit using commands like
audit_accessibility,analyze_bundle, orscan_components. - Review the analysis to see high-level summaries and tool-specific evidence.
- Implement remediation following the generated actionable steps and fix instructions.
Frameworks & tools
Works with React, TypeScript, Webpack, and Vite. Specifically requires stats.json for bundle analysis and tsconfig.json for component documentation.
Why this beats prompting it yourself
Generic LLMs cannot inspect your actual build artifacts or live DOM structure. This skill integrates directly with specific MCP tools to extract real data from your bundles and source code, providing evidence-based fixes instead of guesses.
Use cases
- Reduce bundle size by finding and removing duplicate package versions in your build.
- Fix accessibility by identifying specific elements missing ARIA labels or proper contrast.
- Automate documentation by generating prop tables for a shared UI library.
- Optimize Core Web Vitals by identifying high-impact performance opportunities for LCP and CLS.
Known limitations
Bundle analysis requires a pre-generated stats.json file. Component documentation works best with .tsx files and a valid tsconfig.json.
Requires stats.json for bundle analysis. Component documentation works best with .tsx files and requires tsconfig.json.
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 any payment.
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
Free forever
No account required to browse
Trust & safety
Security scan
Security scan passed 18 days ago
- Free with an Agensi account
Requires