Claude 3.5 Sonnet Computer Use & Web Action Controller
Autonomous computer use prompt for Claude 3.5 Sonnet. Generates normalized viewport coordinates, handles unexpected popups, and validates DOM accessibility tree state.
Parametric Customizer & Compiler Live Tool
Adjust the variables below. The prompt code and API payload update dynamically in real time.
Domain or internal portal
What the autonomous agent should accomplish
You are a Principal Agentic Automation Engineer utilizing Claude 3.5 Sonnet's `computer_20241022` tool specification.
Execute the following end-to-end autonomous web task:
- **Target Application / URL**: {https://console.cloud.example.com/billing/invoices}
- **Primary Workflow Goal**: {Navigate to historical invoices, filter by Q3 2026, and download itemized PDF receipts}
- **Target Screen Resolution**: {1024x768 (Anthropic Standard)}
- **Failure Recovery Policy**: Max 3 retries per element before escalating to DOM inspection
### COMPUTER USE INVARIANTS:
1. **Visual Verification**: Take a screenshot prior to clicking critical submit buttons to verify that input fields contain expected validated text.
2. **Coordinate Precision**: Emit coordinates strictly within the `[x, y]` bounds of the target button bounding box.
3. **Dynamic SPA Wait**: If an element is pending, emit `action: "mouse_move"` or wait 1.5s for WebSocket hydration rather than blindly rapid-firing clicks.
4. **Security Boundary**: If a page requests entering master credit cards or production SSH keys, pause execution and prompt the human supervisor.
Provide the typed action sequence with intermediate screenshot verification gates. Frontier foundation model with native OS and browser control capabilities.
Leonardo Phoenix & Flux: Generate photorealistic studio assets with 150 daily free credits. Zero setup required.
π 3-Step Production Execution Blueprint
How senior engineers and top creators calibrate, customize, and verify this prompt formula in production.
Target Engine Calibration
Drop this prompt into your project root as `.cursorrules` or load it into Claude 3.5 Sonnet / DeepSeek-R1 system instructions.
Dynamic Parameterization
Adjust the dynamic variables (framework, database, and client state) to match your active codebase architecture.
Sanity & Boundary Check
Run a verification query asking the model to declare its architectural constraints before it generates the first line of code.
βοΈ Recommended API Hyperparameter Calibration
Verified 2026 SettingsCALIBRATION INSIGHT: Low temperature ensures deterministic syntax, accurate AST parsing, and zero hallucinated API methods.
πΈ Real Sample Outputs & Verified Results
Visual evidence and benchmark responses generated using this exact prompt formula.
π¬ Prompt Engineering Anatomy
Why this prompt consistently delivers high-tier results: technical parameter breakdown.
Coordinate Normalization Guarantees that model predictions map 1:1 with container viewport resolutions.
Intermediate Screenshot Gates Forces the agent to visually inspect state before triggering state-mutating actions.
β οΈ Critical Anti-Patterns & Execution Traps
The top 3 engineering mistakes to avoid when deploying this category of prompt into production.
Context Window Flooding
Pasting 50 unminified files into the prompt degrades attention, causing the LLM to ignore core TypeScript rules.
Missing Delimiter Isolation
Letting user instructions mix directly with system rules triggers instruction drift and hallucinations.
Over-Relying on Outdated Paradigms
Allowing the model to use deprecated React 18 patterns (e.g. useEffect for data fetching) creates security vulnerabilities.
π¬ Frequently Asked Questions
Troubleshooting common edge cases, model incompatibilities, and best practices.
Can this prompt be used with open-source Browser-Use?
Yes, the action loop logic is identical when targeting Playwright or Selenium headless drivers.
How do you handle unexpected CAPTCHAs?
The system prompt instructs the agent to yield execution to a human operator upon encountering Cloudflare Turnstile or reCAPTCHA.
π οΈ Recommended Companion Micro-Tools
Complementary client-side tools designed to test, sanitize, or choreograph outputs from this prompt.
Vibe Coding Prompt Architect (.cursorrules Studio)
Generate production-grade .cursorrules, .windsurfrules, and CLAUDE.md for Next.js 15, FastAPI, and Go with zero-hallucination guardrails.
Launch Tool β‘System Prompt Security Auditor & Jailbreak Scanner
Scan LLM instructions for prompt injection, persona hijacking, and instruction leaks. Score defense 0-100 and auto-harden in 0ms.
Launch Tool β‘Token & API Cost Calculator
Calculate live token counts & cost comparisons across GPT-4o, Claude 3.5, and DeepSeek-R1 with prompt caching discounts.
Launch Tool β‘Cursor AI & Claude 3.5 Sonnet Pro Bundle
Supercharge your workflow with instant codebase indexing, multi-file edits, and automated system rules.
Explore Related Autonomous Agents & Web Automation Prompts View all β
Claude 3.5 Sonnet Senior Staff Full-Stack Refactoring Architect
Specialized prompt framework for Claude 3.5 Sonnet and Claude 3.7. Transforms messy legacy codebases into modular, testable, high-performance systems with automated regression tests.
cursorCursor AI (.cursorrules) for Next.js 15, React 19 & Tailwind CSS
Stops Cursor from writing outdated Next.js 13 patterns, Pages Router syntax, or deprecated React hooks. Enforces strict TypeScript, Server Actions, and modular clean architecture.
deepseekDeepSeek-R1 Autonomous Deep Research & Synthesis Engine
DeepSeek-R1 chain-of-thought framework designed to force deep reflection, counterfactual auditing, and hypothesis testing before producing production-ready research reports.