{"slug": "show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs", "title": "Show HN: QAgent – a self-healing QA agent that fixes bugs and opens PRs", "summary": "QAgent, a self-healing QA agent that automatically tests web applications, identifies bugs, applies fixes, and verifies fixes without human intervention, has been released as an open-source project on GitHub. The multi-agent system uses Browserbase and Stagehand for AI-powered browser automation, Redis as a vector knowledge base for learning from past bugs, and Vercel for instant deployment after fixes, with the goal of creating a closed-loop for automated bug detection and fixing.", "body_md": "A self-improving QA agent that automatically tests web applications, identifies bugs, applies fixes, and verifies the fixes – all without human intervention.\n\nQAgent is a multi-agent system that creates a closed-loop for automated bug detection and fixing:\n\n```\n┌──────────┐    ┌──────────┐    ┌──────────┐    ┌──────────┐\n│  TESTER  │───▶│  TRIAGE  │───▶│  FIXER   │───▶│ VERIFIER │\n│  Agent   │    │  Agent   │    │  Agent   │    │  Agent   │\n└──────────┘    └──────────┘    └──────────┘    └──────────┘\n     │                                               │\n     │              ┌──────────────┐                 │\n     │              │    Redis     │◀────────────────┘\n     │              │ (Knowledge   │\n     │              │    Base)     │\n     │              └──────────────┘\n     │                    │\n     ▼                    ▼\n┌─────────────────────────────────────────────────────────┐\n│              W&B Weave (Observability)                  │\n└─────────────────────────────────────────────────────────┘\n```\n\n**Continuous Testing**: Runs E2E tests like a QA engineer, simulating real user flows** Automatic Bug Fixing**: Doesn't just report bugs – it fixes them and redeploys** Self-Improvement**: Learns from past bugs to diagnose and fix faster over time** Measurable Impact**: Track pass rates, time-to-fix, and iterations to prove improvement\n\n```\n# Clone the repository\ngit clone https://github.com/rishabhcli/QAgent.git\ncd QAgent\n\n# Install dependencies\npnpm install\n\n# Set up environment\ncp .env.example .env.local\n# Edit .env.local with your API keys\n\n# Run development server (demo app)\npnpm dev\n\n# Run the QAgent agent\npnpm run agent\n\n# Start the Marimo dashboard\nmarimo run dashboard/app.py\n```\n\n| Technology | Purpose |\n|---|---|\nBrowserbase + Stagehand |\nAI-powered browser automation for E2E testing |\nVercel |\nInstant deployment after fixes |\nRedis |\nVector knowledge base for learning from past bugs |\nW&B Weave |\nTracing and evaluation of agent runs |\nCustom Orchestrator (ADK/A2A-compatible) |\nMulti-agent workflow coordination (ADK integration planned) |\nMarimo |\nInteractive analytics dashboard |\nNext.js |\nDemo application |\nOpenAI |\nLLM for patch generation |\n\n| File | Purpose |\n|---|---|\n|\n\n[TASKS.md](/rishabhcli/QAgent/blob/main/TASKS.md)[docs/PRD.md](/rishabhcli/QAgent/blob/main/docs/PRD.md)[docs/DESIGN.md](/rishabhcli/QAgent/blob/main/docs/DESIGN.md)[docs/ARCHITECTURE.md](/rishabhcli/QAgent/blob/main/docs/ARCHITECTURE.md)[prompts/ralph-loop.md](/rishabhcli/QAgent/blob/main/prompts/ralph-loop.md)\n\n```\nQAgent/\n├── .claude/\n│   └── skills/               # Domain-specific knowledge modules\n│       ├── browserbase-stagehand/\n│       ├── redis-vectorstore/\n│       ├── vercel-deployment/\n│       ├── wandb-weave/\n│       ├── google-adk/\n│       ├── marimo-dashboards/\n│       └── qagent-agents/\n├── agents/                   # Agent implementations\n│   ├── tester/\n│   ├── triage/\n│   ├── fixer/\n│   ├── verifier/\n│   └── orchestrator/\n├── app/                      # Next.js demo app\n├── dashboard/                # Marimo analytics\n├── docs/                     # Documentation\n├── lib/                      # Shared libraries\n├── prompts/                  # Workflow prompts\n└── tests/                    # Test suites\n```\n\n**Test**- Tester Agent runs E2E tests using Browserbase/Stagehand** Detect**- Failures are captured with screenshots, DOM state, logs** Diagnose**- Triage Agent analyzes the failure and queries Redis for similar issues** Fix**- Fixer Agent generates a patch using LLM + past fix patterns** Deploy**- Verifier Agent applies the patch and deploys via Vercel** Verify**- Tests are re-run to confirm the fix works** Learn**- Successful fixes are stored in Redis for future reference** Repeat**- Loop continues until all tests pass\n\n**Knowledge Base**: Every bug and fix is stored with embeddings for semantic search** Pattern Learning**: Similar bugs are fixed faster using past solutions** TraceTriage**: Agent failures are analyzed to improve prompts and workflows** RedTeam**: Adversarial tests continuously harden the system\n\n**Start every session** by reading[CLAUDE.md](/rishabhcli/QAgent/blob/main/CLAUDE.md)**Check current work** in[TASKS.md](/rishabhcli/QAgent/blob/main/TASKS.md)**Follow the Ralph Loop** workflow for iterative development**Load skills** from`.claude/skills/`\n\nas needed\n\n```\n# Install dependencies\npnpm install\n\n# Run demo app\npnpm dev\n\n# Run agent\npnpm run agent\n\n# Run tests\npnpm test\n\n# Run E2E tests\npnpm run test:e2e\n\n# Lint and format\npnpm lint && pnpm format\n\n# Build\npnpm build\n```\n\nSee [.env.example](/rishabhcli/QAgent/blob/main/.env.example) for required environment variables:\n\n`BROWSERBASE_API_KEY`\n\n- Browserbase API key`OPENAI_API_KEY`\n\n- OpenAI API key`REDIS_URL`\n\n- Redis connection string`VERCEL_TOKEN`\n\n- Vercel API token`WANDB_API_KEY`\n\n- Weights & Biases API key`GOOGLE_CLOUD_PROJECT`\n\n- Google Cloud project (reserved for ADK/A2A integration)\n\nSee the [Quick Start](#quick-start) section above for setup instructions. Once running, connect a GitHub repository through the dashboard and start your first QAgent run.\n\n[QAgent Paper](https://arxiv.org/html/2502.02747v1)- Agentic patching framework[Stagehand](https://www.stagehand.dev/)- AI browser automation[Browserbase](https://browserbase.com/)- Cloud browsers[W&B Weave](https://wandb.ai/site/weave)- LLM observability[Google ADK](https://cloud.google.com/agent-development-kit)- Planned orchestration framework[Marimo](https://marimo.io/)- Reactive notebooks", "url": "https://wpnews.pro/news/show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs", "canonical_source": "https://github.com/rishabhcli/QAgent", "published_at": "2026-07-22 13:30:45+00:00", "updated_at": "2026-07-22 13:52:45.826505+00:00", "lang": "en", "topics": ["ai-agents", "ai-tools", "ai-research", "developer-tools"], "entities": ["QAgent", "Browserbase", "Stagehand", "Redis", "Vercel", "W&B Weave", "OpenAI", "Next.js"], "alternates": {"html": "https://wpnews.pro/news/show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs", "markdown": "https://wpnews.pro/news/show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs.md", "text": "https://wpnews.pro/news/show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs.txt", "jsonld": "https://wpnews.pro/news/show-hn-qagent-a-self-healing-qa-agent-that-fixes-bugs-and-opens-prs.jsonld"}}