Launch a Chrome/Edge instance with automatic WebGPU Inspector instrumentation. Optionally open a URL in the instrumented browser for immediate inspection.
-
523,639 tools. Updated 2026-09-06 13:38
"WebGPU" matching MCP tools:
- Load a saved WebGPU capture (.wgpuc or .json) from disk to inspect frames, shaders, and buffers for performance and rendering analysis.-
- Retrieve WebGPU validation errors from a captured frame to identify and fix rendering issues.-
- Update PlayCanvas project settings (physics, rendering, layers, input, resolution) with partial path/value pairs; returns the full resulting settings snapshot.MIT
- Evaluate JavaScript in the play page's top frame for browser diagnostics: WebGPU support, network state, document title. Can't reach game UI; use spawn_play_input and spawn_exec.MIT
- Need to see and interact with a live Spawn game? Open its play URL in a headed Chromium window via Playwright, then use screenshots and input to control it. Returns WebGPU availability.MIT
Matching MCP Servers
- FlicenseAqualityBmaintenanceEnables AI assistants to drive Chrome/Chromium to capture live WebGPU frames, profile performance, inspect shaders, buffers, and textures, and diagnose rendering issues.21-
- AlicenseAqualityDmaintenanceEnables AI agents to control and manipulate live 3D scenes across frameworks like Three.js, A-Frame, and Babylon.js using a comprehensive set of object and environment tools. It features an integrated in-world chat system that allows for real-time scene modifications directly from within the 3D canvas.33523MIT
- Inspect a live GPU texture or render target region without full capture, returning per-channel stats, hole coverage, and an ASCII luminance view. Resolves texture views automatically; use when a capture is armed.-
- Determines the correct rendering fallback for Studio surfaces by returning a JSON contract of renderer capabilities.-
- Capture a PNG screenshot of the composited page to visually verify WebGPU rendering results as presented, independent of render target pooling.-
- Read live GPU buffer contents directly from a connected page, decoding bytes as uint8, float32, or other types, without needing a full capture.-
- Sample live frame-health metrics to detect dropped frames and CPU-bound rendering, returning FPS, frame times, and a CPU vs GPU/vsync bound verdict.-
- Lists browser pages currently connected to the live WebGPU bridge, indicating which are instrumented and ready for inspection.-
- Summarize a WebGPU capture to diagnose performance and correctness issues, providing object counts, command counts, shader entry points, validation errors, and heuristic findings.-
- Decode captured vertex buffer data into per-attribute values like UV coordinates, automatically using the draw's pipeline layout from a setVertexBuffer command index.-
- Identify CPU, GPU, or vsync-bound performance in a WebGPU capture and get ranked render passes with bottleneck analysis and concrete improvement suggestions.-
- Resolve the full GPU state of any draw/dispatch command: bound pipeline, bind groups, vertex buffers, index buffer, and draw params. Diagnose what a specific draw read.-
- Opens a new instrumented browser tab, navigates to a specified URL, and waits for the page to connect, returning it ready for WebGPU frame capture and profiling.-
- Check browser connection status and view connection method, instrumented targets, and pages connected to the bridge for WebGPU debugging.-
- Retrieve paginated command lists from a WebGPU capture, filtered by method or pass label, with index, arguments, and object details.-
- Identify differences in resolved pipeline, bind groups, and bindings between two draw commands to pinpoint why one draw renders correctly and another does not.-
- List all live and loaded WebGPU captures available for analysis, enabling you to select one for performance profiling and rendering diagnostics.-