Home - News ( United States | United Kingdom | Italy | Germany ) - Football scores

Showing content from https://docs.anthropic.com/en/docs/agents-and-tools/tool-use/code-execution-tool below:

Code execution tool - Anthropic

The code execution tool allows Claude to execute Python code in a secure, sandboxed environment. Claude can analyze data, create visualizations, perform complex calculations, and process uploaded files directly within the API conversation.

The code execution tool is available on:

• Claude Opus 4.1 (claude-opus-4-1-20250805)
• Claude Opus 4 (claude-opus-4-20250514)
• Claude Sonnet 4 (claude-sonnet-4-20250514)
• Claude Sonnet 3.7 (claude-3-7-sonnet-20250219)
• Claude Haiku 3.5 (claude-3-5-haiku-latest)

Here’s a simple example that asks Claude to perform a calculation:

When you add the code execution tool to your API request:

1. Claude evaluates whether code execution would help answer your question
2. Claude writes and executes Python code in a secure sandbox environment
3. Code execution may occur multiple times throughout a single request
4. Claude provides results with any generated charts, calculations, or analysis

The code execution tool requires no additional parameters:

Here’s an example response with code execution:

Code execution results include:

• stdout: Output from print statements and successful execution
• stderr: Error messages if code execution fails
• return_code (0 for success, non-zero for failure)

If there is an error using the tool there will be a code_execution_tool_result_error

Possible errors include

• unavailable: The code execution tool is unavailable
• code_execution_exceeded: Execution time exceeded the maximum allowed
• container_expired: The container is expired and not available

The response may include a pause_turn stop reason, which indicates that the API paused a long-running turn. You may provide the response back as-is in a subsequent request to let Claude continue its turn, or modify the content if you wish to interrupt the conversation.

Code execution can analyze files uploaded via the Files API, such as CSV files, Excel files, and other data formats. This allows Claude to read, process, and generate insights from your data. You can pass multiple files per request.

The Python environment is capable of working with but not limited to the following file types

• CSV
• Excel (.xlsx, .xls)
• JSON
• XML
• Images (JPEG, PNG, GIF, WebP)
• Text files (.txt, .md, .py, etc)

1. Upload your file using the Files API
2. Reference the file in your message using a container_upload content block
3. Include the code execution tool in your API request

When Claude creates files during code execution (e.g., saving matplotlib plots, generating CSVs), you can retrieve these files using the Files API:

The code execution tool runs in a secure, containerized environment designed specifically for Python code execution.

• Python version: 3.11.12
• Operating system: Linux-based container
• Architecture: x86_64 (AMD64)

• Memory: 1GiB RAM
• Disk space: 5GiB workspace storage
• CPU: 1 CPU

• Internet access: Completely disabled for security
• External connections: No outbound network requests permitted
• Sandbox isolation: Full isolation from host system and other containers
• File access: Limited to workspace directory only
• Workspace scoping: Like Files, containers are scoped to the workspace of the API key
• Expiration: Containers expire 1 hour after creation

The sandboxed Python environment includes these commonly used libraries:

• Data Science: pandas, numpy, scipy, scikit-learn, statsmodels
• Visualization: matplotlib
• File Processing: pyarrow, openpyxl, xlrd, pillow
• Math & Computing: sympy, mpmath
• Utilities: tqdm, python-dateutil, pytz, joblib

You can reuse an existing container across multiple API requests by providing the container ID from a previous response. This allows you to maintain created files between requests.

With streaming enabled, you’ll receive code execution events as they occur:

You can include the code execution tool in the Messages Batches API. Code execution tool calls through the Messages Batches API are priced the same as those in regular Messages API requests.

The code execution tool usage is tracked separately from token usage. Execution time is a minimum of 5 minutes. If files are included in the request, execution time is billed even if the tool is not used due to files being preloaded onto the container.

Pricing: $0.05 per session-hour.

RetroSearch is an open source project built by @garambo | Open a GitHub Issue

Search and Browse the WWW like it's 1997 | Search results from DuckDuckGo

HTML: 3.2 | Encoding: UTF-8 | Version: 0.7.4