Frontend Performance Lab MCP
Connect your AI assistant to a specialized frontend performance knowledge graph. Use Frontend Performance Lab from Cursor, Claude Desktop, VS Code, Windsurf and any MCP-compatible client.
Live Server Status
Real-time diagnostics from the running MCP API endpoint.
What can Frontend Performance Lab do?
Review Component Performance
Analyze Vue and React code for hidden performance issues.
Detect Rendering Bottlenecks
Identify layout thrashing and large component trees instantly.
Explain Browser APIs
Get structured documentation on performance APIs directly in your IDE.
Recommend Optimizations
Provide actionable recipes to fix identified bottlenecks.
Search Performance Knowledge
Query a curated database of modern frontend patterns.
Launch Interactive Experiments
Open live, sandboxed demonstrations of specific bugs.
Architecture
How the Frontend Performance Lab integrates with your IDE.
Example Workflow
The automated "aha" moment powered by MCP tools.
Try Frontend Performance Lab API Powered
Execute real MCP tools directly from your browser. This playground communicates with the same unified core as the stdio server via the Nuxt /api/mcp/execute endpoint.
// Output will appear here...
Tools Explorer
list_experiments
KnowledgeList all available frontend performance experiments.
get_experiment
KnowledgeGet full manifest or specific section.
list_browser_apis
Browser APIsList all available Browser APIs in the performance registry.
get_browser_api
Browser APIsGet the full details of a specific Browser API.
list_recipes
KnowledgeList all available performance recipes.
get_recipe
KnowledgeGet the full details of a specific recipe.
search
SearchSearch the knowledge base.
system_diagnostics
DiagnosticsGenerate system health report.
Resources Explorer
Experiments
performance://experiments/{id}Browser APIs
performance://browser-apis/{id}Recipes
performance://recipes/{id}Checklists
performance://checklists/{id}Prompt Explorer
Review Performance
Tool: review_performance
Ask the AI to review your current code for performance bottlenecks based on the lab guidelines.