René's URL Explorer Experiment


Title: Computer use tool - Claude Platform Docs

Open Graph Title: Computer use tool

X Title: Computer use tool

Description: Claude API Documentation

Open Graph Description: Claude API Documentation

X Description: Claude API Documentation

Opengraph URL: https://platform.claude.com/docs/en/agents-and-tools/tool-use/computer-use-tool

direct link

Domain: docs.anthropic.com

mobile-web-app-capableyes
apple-mobile-web-app-status-bar-styledefault
og:site_nameClaude Platform Docs
og:localeen
og:imagehttps://platform.claude.com/docs/og?locale=en&path=agents-and-tools/tool-use/computer-use-tool&design-rev=1
og:image:width1200
og:image:height630
og:image:altComputer use tool
og:image:typeimage/png
og:typearticle
twitter:cardsummary_large_image
twitter:imagehttps://platform.claude.com/docs/og?locale=en&path=agents-and-tools/tool-use/computer-use-tool&design-rev=1

Links:

Claude Platform Docshttps://docs.anthropic.com/docs/en/home
Messageshttps://docs.anthropic.com/docs/en/intro
Managed Agentshttps://docs.anthropic.com/docs/en/managed-agents/overview
Adminhttps://docs.anthropic.com/docs/en/manage-claude/admin-api
API referencehttps://docs.anthropic.com/docs/en/api/overview
Consolehttps://docs.anthropic.com/
Log inhttps://docs.anthropic.com/login?returnTo=%2Fdocs%2Fen%2Fagents-and-tools%2Ftool-use%2Fcomputer-use-tool
Intro to Claudehttps://docs.anthropic.com/docs/en/intro
Quickstarthttps://docs.anthropic.com/docs/en/get-started
Features overviewhttps://docs.anthropic.com/docs/en/build-with-claude/overview
Using the Messages APIhttps://docs.anthropic.com/docs/en/build-with-claude/working-with-messages
Stop reasons and fallbackhttps://docs.anthropic.com/docs/en/build-with-claude/handling-stop-reasons
Refusals and fallbackhttps://docs.anthropic.com/docs/en/build-with-claude/refusals-and-fallback
Fallback credithttps://docs.anthropic.com/docs/en/build-with-claude/fallback-credit
Extended thinkinghttps://docs.anthropic.com/docs/en/build-with-claude/extended-thinking
Adaptive thinkinghttps://docs.anthropic.com/docs/en/build-with-claude/adaptive-thinking
Efforthttps://docs.anthropic.com/docs/en/build-with-claude/effort
Task budgets (beta)https://docs.anthropic.com/docs/en/build-with-claude/task-budgets
Fast mode (research preview)https://docs.anthropic.com/docs/en/build-with-claude/fast-mode
Structured outputshttps://docs.anthropic.com/docs/en/build-with-claude/structured-outputs
Citationshttps://docs.anthropic.com/docs/en/build-with-claude/citations
Streaming Messageshttps://docs.anthropic.com/docs/en/build-with-claude/streaming
Batch processinghttps://docs.anthropic.com/docs/en/build-with-claude/batch-processing
Search resultshttps://docs.anthropic.com/docs/en/build-with-claude/search-results
Streaming refusalshttps://docs.anthropic.com/docs/en/test-and-evaluate/strengthen-guardrails/handle-streaming-refusals
Multilingual supporthttps://docs.anthropic.com/docs/en/build-with-claude/multilingual-support
Embeddingshttps://docs.anthropic.com/docs/en/build-with-claude/embeddings
Overviewhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/overview
How tool use workshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/how-tool-use-works
Tutorial: Build a tool-using agenthttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/build-a-tool-using-agent
Define toolshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/define-tools
Handle tool callshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/handle-tool-calls
Parallel tool usehttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/parallel-tool-use
Tool Runner (SDK)https://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-runner
Strict tool usehttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/strict-tool-use
Server toolshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/server-tools
Web search toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/web-search-tool
Web fetch toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/web-fetch-tool
Code execution toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/code-execution-tool
Advisor toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/advisor-tool
Tool search toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-search-tool
Memory toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/memory-tool
Bash toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/bash-tool
Text editor toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/text-editor-tool
Computer use toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/computer-use-tool
Troubleshootinghttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/troubleshooting-tool-use
Tool referencehttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-reference
Manage tool contexthttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/manage-tool-context
Tool combinationshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-combinations
Tool use with prompt cachinghttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-use-with-prompt-caching
Programmatic tool callinghttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/programmatic-tool-calling
Fine-grained tool streaminghttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/fine-grained-tool-streaming
Context windowshttps://docs.anthropic.com/docs/en/build-with-claude/context-windows
Compactionhttps://docs.anthropic.com/docs/en/build-with-claude/compaction
Context editinghttps://docs.anthropic.com/docs/en/build-with-claude/context-editing
Prompt cachinghttps://docs.anthropic.com/docs/en/build-with-claude/prompt-caching
Mid-conversation system messageshttps://docs.anthropic.com/docs/en/build-with-claude/mid-conversation-system-messages
Build an orchestration modehttps://docs.anthropic.com/docs/en/build-with-claude/mid-conversation-effort-example
Cache diagnostics (beta)https://docs.anthropic.com/docs/en/build-with-claude/cache-diagnostics
Token countinghttps://docs.anthropic.com/docs/en/build-with-claude/token-counting
Files APIhttps://docs.anthropic.com/docs/en/build-with-claude/files
PDF supporthttps://docs.anthropic.com/docs/en/build-with-claude/pdf-support
Overviewhttps://docs.anthropic.com/docs/en/agents-and-tools/agent-skills/overview
Quickstarthttps://docs.anthropic.com/docs/en/agents-and-tools/agent-skills/quickstart
Best practiceshttps://docs.anthropic.com/docs/en/agents-and-tools/agent-skills/best-practices
Skills for enterprisehttps://docs.anthropic.com/docs/en/agents-and-tools/agent-skills/enterprise
Skills in the APIhttps://docs.anthropic.com/docs/en/build-with-claude/skills-guide
Remote MCP servershttps://docs.anthropic.com/docs/en/agents-and-tools/remote-mcp-servers
MCP connectorhttps://docs.anthropic.com/docs/en/agents-and-tools/mcp-connector
Amazon Bedrockhttps://docs.anthropic.com/docs/en/build-with-claude/claude-in-amazon-bedrock
Amazon Bedrock (legacy)https://docs.anthropic.com/docs/en/build-with-claude/claude-on-amazon-bedrock-legacy
Claude Platform on AWShttps://docs.anthropic.com/docs/en/build-with-claude/claude-platform-on-aws
Google Cloudhttps://docs.anthropic.com/docs/en/build-with-claude/claude-on-vertex-ai
Microsoft Foundryhttps://docs.anthropic.com/docs/en/build-with-claude/claude-in-microsoft-foundry
Log inhttps://docs.anthropic.com/login
Claude Platform Docshttps://docs.anthropic.com/docs
https://x.com/claudeai
https://www.threads.com/@claudeai
https://www.linkedin.com/showcase/claude
https://www.youtube.com/@anthropic-ai
https://instagram.com/claudeai
AI agentshttps://claude.com/solutions/agents
Code modernizationhttps://claude.com/solutions/code-modernization
Codinghttps://claude.com/solutions/coding
Customer supporthttps://claude.com/solutions/customer-support
Educationhttps://claude.com/solutions/education
Financial serviceshttps://claude.com/solutions/financial-services
Governmenthttps://claude.com/solutions/government
Life scienceshttps://claude.com/solutions/life-sciences
Claude on AWShttps://claude.com/partners/amazon-bedrock
Claude on Google Cloudhttps://claude.com/partners/google-cloud-vertex-ai
Bloghttps://claude.com/blog
Courseshttps://claude.com/resources/courses
Use caseshttps://claude.com/resources/use-cases
Connectorshttps://claude.com/partners/mcp
Customer storieshttps://claude.com/customers
Engineering at Anthropichttps://www.anthropic.com/engineering
Eventshttps://www.anthropic.com/events
Powered by Claudehttps://claude.com/partners/powered-by-claude
Service partnershttps://claude.com/partners/services
Startups programhttps://claude.com/programs/startups
Anthropichttps://www.anthropic.com/company
Careershttps://www.anthropic.com/careers
Economic Futureshttps://www.anthropic.com/economic-futures
Researchhttps://www.anthropic.com/research
Newshttps://www.anthropic.com/news
Responsible Scaling Policyhttps://www.anthropic.com/news/announcing-our-updated-responsible-scaling-policy
Security and compliancehttps://trust.anthropic.com
Transparencyhttps://www.anthropic.com/transparency
Bloghttps://claude.com/blog
Courseshttps://claude.com/resources/courses
Use caseshttps://claude.com/resources/use-cases
Connectorshttps://claude.com/partners/mcp
Customer storieshttps://claude.com/customers
Engineering at Anthropichttps://www.anthropic.com/engineering
Eventshttps://www.anthropic.com/events
Powered by Claudehttps://claude.com/partners/powered-by-claude
Service partnershttps://claude.com/partners/services
Startups programhttps://claude.com/programs/startups
Availabilityhttps://www.anthropic.com/supported-countries
Statushttps://status.claude.com/
Supporthttps://support.claude.com/
Discordhttps://www.anthropic.com/discord
Privacy policyhttps://www.anthropic.com/legal/privacy
Responsible disclosure policyhttps://www.anthropic.com/responsible-disclosure-policy
Terms of service: Commercialhttps://www.anthropic.com/legal/commercial-terms
Terms of service: Consumerhttps://www.anthropic.com/legal/consumer-terms
Usage policyhttps://www.anthropic.com/legal/aup
WebArenahttps://webarena.dev/
beta headerhttps://docs.anthropic.com/docs/en/api/beta-headers
deprecatedhttps://docs.anthropic.com/docs/en/about-claude/model-deprecations
retired, except on Bedrock and Google Cloudhttps://docs.anthropic.com/docs/en/about-claude/model-deprecations
retired, except on Google Cloudhttps://docs.anthropic.com/docs/en/about-claude/model-deprecations
feedback formhttps://forms.gle/H6UFuXaaLywri9hz6
Zero Data Retention (ZDR)https://docs.anthropic.com/docs/en/build-with-claude/api-and-data-retention
Tool referencehttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/tool-reference
contact supporthttps://support.claude.com/en/
Computer use reference implementationGet started with the computer use reference implementation that includes a web interface, Docker container, example tool implementations, and an agent loop.https://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo
reference implementationhttps://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo
containerized environmenthttps://github.com/anthropics/anthropic-quickstarts/blob/main/computer-use-demo/Dockerfile
the computer use toolshttps://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo/computer_use_demo/tools
agent loophttps://github.com/anthropics/anthropic-quickstarts/blob/main/computer-use-demo/computer_use_demo/loop.py
Mitigate jailbreaks and prompt injectionshttps://docs.anthropic.com/docs/en/test-and-evaluate/strengthen-guardrails/mitigate-jailbreaks
Effective harnesses for long-running agentshttps://www.anthropic.com/engineering/effective-harnesses-for-long-running-agents
tool use system prompthttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/define-tools#tool-use-system-prompt
Extended thinkinghttps://docs.anthropic.com/docs/en/build-with-claude/extended-thinking
Quick starthttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#quick-start
bash toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/bash-tool
text editor toolhttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/text-editor-tool
custom tool definitionshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/define-tools
reference implementationhttps://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo
image size limitshttps://docs.anthropic.com/docs/en/build-with-claude/vision#evaluate-image-size
API and data retentionhttps://docs.anthropic.com/docs/en/manage-claude/api-and-data-retention
tool use pricinghttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/overview#pricing
Vision pricinghttps://docs.anthropic.com/docs/en/build-with-claude/vision
Text editor toolContinue to the next tool: view, create, and edit files with Claudehttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/text-editor-tool
Reference implementationGet started with the complete Docker-based implementationhttps://github.com/anthropics/anthropic-quickstarts/tree/main/computer-use-demo
Tool documentationLearn more about tool use and creating custom toolshttps://docs.anthropic.com/docs/en/agents-and-tools/tool-use/overview
Best practices in detailBenchmarked recommendations for resolution, thinking effort, and context managementhttps://claude.com/blog/best-practices-for-computer-and-browser-use-with-claude
Overviewhttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#overview
Security considerationshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#security-considerations
Quick starthttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#quick-start
How computer use workshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#how-computer-use-works
The computing environmenthttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#the-computing-environment
How to implement computer usehttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#how-to-implement-computer-use
Start with the reference implementationhttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#start-with-the-reference-implementation
Understanding the agentic loophttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#understanding-the-agentic-loop
Optimize model performance with promptinghttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#optimize-model-performance-with-prompting
System promptshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#system-prompts
Available actionshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#available-actions
Tool parametershttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#tool-parameters
Combining with extended thinkinghttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#combining-with-extended-thinking
Augmenting computer use with other toolshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#augmenting-computer-use-with-other-tools
Build a custom computer use environmenthttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#build-a-custom-computer-use-environment
Understand computer use limitationshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#understand-computer-use-limitations
Data retentionhttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#data-retention
Pricinghttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#pricing
Next stepshttps://docs.anthropic.com/en/docs/agents-and-tools/computer-use#next-steps

Viewport: width=device-width, initial-scale=1, maximum-scale=1, viewport-fit=cover

URLs of crawlers that visited me.