code_interpreter_toolkit

Build a session key from RunnableConfig.

Uses thread_id as the base key and appends checkpoint_ns when present. This ensures parallel LangGraph subgraphs get separate sessions even when they share a thread_id.

Examples::

Top-level agent:   "thread-001"
Subagent A:        "thread-001:research-acme:abc123"
Subagent B:        "thread-001:research-beta:def456"

Create and setup a CodeInterpreterToolkit.

Input schema for execute_code tool.

Input schema for execute_command tool.

Input schema for read_files tool.

Input schema for write_files tool.

Input schema for list_files tool.

Input schema for delete_files tool.

Input schema for upload_file tool.

Input schema for install_packages tool.

Toolkit for working with AWS code interpreter environment.

This toolkit provides a set of tools for working with a remote code interpreter environment:

• execute_code - Run code in various languages (primarily Python)
• execute_command - Run shell commands
• read_files - Read content of files in the environment
• list_files - List files in directories
• delete_files - Remove files from the environment
• write_files - Create or update files
• start_command_execution - Start long-running commands asynchronously
• get_task - Check status of async tasks
• stop_task - Stop running tasks
• upload_file - Upload files with semantic descriptions
• install_packages - Install Python packages

The toolkit lazily initializes the code interpreter session on first use. It supports multiple threads by maintaining separate code interpreter sessions for each thread ID.