GAPS QA
Enables UI auditing via browser automation, capturing per-element evidence and generating Markdown, JSON, and DOCX reports.
README
GAPS QA

GAPS QA is an evidence-first UI-auditing system that turns exploratory browser runs into durable, reviewable engineering artifacts. It combines a React operator console, a remote MCP surface, a Playwright/CDP runner, run persistence, live progress streaming, and deterministic Markdown, JSON, and DOCX reporting.
The design goal is not merely to say that a page “looks wrong.” Every completed audit can preserve per-element findings, machine-readable evidence, human-readable write-ups, and a canonical bundle that another engineer or agent can inspect independently.
Why it stands out
- Browser automation is isolated behind a runner contract; Playwright over CDP is the active adapter.
- The Node orchestration service exposes HTTP, SSE, bundle-serving, and MCP interfaces from one canonical run model.
- Each run emits structured element evidence plus
UI_Element_Writeups.md,UI_Element_Writeups.docx, andui-element-audit.json. - DOCX generation is self-contained and uses Python's standard library rather than a machine-global document package.
- Cloudflare is treated as an edge proxy while the stateful Node service remains the runtime authority.
- The release passes full TypeScript validation, 7 tests including a live browser fixture, and both web and server builds.
flowchart LR
User["Human or ChatGPT client"] --> MCP["MCP + HTTP API"]
User --> Console["React operator console"]
MCP --> Orchestrator["Audit orchestration service"]
Console --> Orchestrator
Orchestrator --> Runner["Playwright / CDP runner"]
Runner --> Target["Audited web application"]
Runner --> Evidence["Per-element evidence"]
Evidence --> Bundle["Markdown + JSON + DOCX bundle"]
Orchestrator --> SSE["Live progress stream"]
Repository map
shared/— canonical schemas, report contracts, and bundle generationrunners/— browser-runner adapters and Playwright/CDP implementationserver/— orchestration, persistence, MCP, HTTP API, SSE, and bundle servingweb/— React operator console and ChatGPT widget surfacepolicy/— single-sourced QA operating contractcloudflare/— optional edge proxy adapter
Quick start
npm ci
npm run check
npm test
npm run build
npm start
Open http://127.0.0.1:3000/app. Completed runs are stored under .data/runs/<run-id>/bundle/ by default, and the MCP endpoint is /mcp.
Deployment shape
render.yaml describes the primary stateful Node deployment. cloudflare/worker.ts and wrangler.jsonc provide an optional edge-proxy layer configured with UPSTREAM_BASE_URL. No deployment URL is advertised until a live smoke check has passed.
Maturity and boundaries
GAPS QA is a substantial working prototype, not a claim of universal visual correctness. Automated evidence still needs human or agent judgment, CDP connectivity must be authorized for the target browser, and deployment operators must add authentication, rate limits, and retention controls appropriate to their environment.
See SECURITY.md, CONTRIBUTING.md, and RIGHTS.md before exposing the service publicly.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.
E2B
Using MCP to run code via e2b.