// Build autonomous AI agents using Claude Agent SDK with computer use, tool calling, MCP integration, and production best practices
| name | Claude SDK Expert |
| description | Build autonomous AI agents using Claude Agent SDK with computer use, tool calling, MCP integration, and production best practices |
| version | 1.1.0 |
| last_updated | "2026-01-06T00:00:00.000Z" |
| external_version | Claude Opus 4.5, Sonnet 4.5 |
| resources | resources/code-examples.py |
Build autonomous AI agents using Claude Agent SDK, leveraging computer use, tool orchestration, and MCP integration for production deployments.
Enables building autonomous agents that control computers, write files, run commands, and iterate on work.
Core Philosophy: Give Claude a computer to unlock agent effectiveness beyond chat.
Claude can control a computer environment:
Use Cases: Finance agents, personal assistants, customer support, development agents, research agents
| Category | Tools |
|---|---|
| Files | Read, Write, Edit |
| Commands | Bash |
| Search | Grep, Glob |
| Web | WebFetch, WebSearch |
Define custom tools via Model Context Protocol servers.
Benefits:
Popular MCP Servers: GitHub, Slack, PostgreSQL, MongoDB, Stripe, Salesforce
Agent completes multi-step task without intervention.
User Request → Analyze → Subtasks → Execute Tools → Iterate → Result
Agent proposes actions, waits for approval.
Task → Plan → Human Review → Approve? → Execute → Result
Agent retries on errors automatically.
Attempt 1 → Error → Analyze → Attempt 2 → Success
See: resources/code-examples.py for full implementations
See: resources/code-examples.py for good/bad tool examples
mcp_config = {
"servers": {
"github": {
"command": "npx",
"args": ["-y", "@modelcontextprotocol/server-github"],
"env": {"GITHUB_TOKEN": os.getenv("GITHUB_TOKEN")}
}
}
}
Full example: resources/code-examples.py
Show real-time progress to build user trust.
Implementation: resources/code-examples.py
| Model | Best For | Pricing (per M tokens) | Speed |
|---|---|---|---|
| claude-opus-4-5 | Flagship reasoning, complex agents, highest accuracy | $5 in / $25 out | Slower |
| claude-sonnet-4-5 | Best balance - coding, agents, computer use | $3 in / $15 out | Medium |
| claude-haiku-4 | Simple tasks, format conversions, high-throughput | $0.25 in / $1.25 out | Fast |
Note: Opus 4.5 achieved 80.9% on SWE-bench Verified. Sonnet 4.5 supports 1M token context with beta header.
Test individual tools in isolation.
Test agent workflows with multiple tools.
Measure accuracy, latency, tool efficiency.
Examples: resources/code-examples.py
| Metric | Description |
|---|---|
| Tool Call Success Rate | % of tool invocations succeeding |
| Task Completion Rate | % of requests fully resolved |
| Average Iterations | Tool calls per task |
| Latency | Time to complete requests |
| Token Usage | Input + output tokens |
| Error Rate | % of requests with errors |
Documentation:
GitHub:
Build powerful, autonomous agents using Claude's cutting-edge capabilities.
Patterns for multi-agent coordination, task decomposition, handoffs, and workflow orchestration. Best practices for building and managing agent systems.
Enterprise AI security - OWASP LLM Top 10, prompt injection defense, guardrails, PII protection
Create professional architecture diagrams using D2, Draw.io, Mermaid, and OCI official icons for enterprise-grade visualizations
Build AI applications on AWS using Bedrock, SageMaker, and AI/ML services with best practices for enterprise deployment
Build AI applications on Azure using Azure OpenAI, Cognitive Services, and ML services with enterprise patterns
Production-grade AI architecture patterns for enterprise - security, governance, scalability, and operational excellence