Skip to main content
In Admin > Settings > Code Execution, configure the runtime for AI-generated code. You can configure two features separately: Code Execution and Code Interpreter.
Code Execution settings tab

Code Execution vs Code Interpreter

Code ExecutionCode Interpreter
BehaviorUser manually runs AI-generated code blocksAI auto-generates and runs code
TriggerClick the “Run” button on a code blockWhen the AI decides data analysis is needed
ActivationThis setting aloneThis setting + agent capability + user chat toggle — all 3 required

Code Execution

A feature that lets users directly run AI-generated code blocks in chat.
SettingDescriptionDefault
Code Execution enabledCode execution feature ON/OFFEnabled
Execution engineCode execution environmentpyodide

Execution Engines

Run Python inside the browser. WebAssembly-based — doesn’t affect the server.
ProCon
No separate server neededSome native packages unsupported
No security riskLarge-data processing limits
Ready to use immediatelyNo filesystem access

Code Interpreter

A feature where the agent auto-generates and runs code for data analysis, chart generation, etc.
SettingDescriptionDefault
Code Interpreter enabledCode Interpreter feature ON/OFFEnabled
Execution engineCode Interpreter runtimepyodide
Prompt templateSystem prompt used when generating code (optional)-
The Code Interpreter also supports Pyodide and Jupyter engines, with Jupyter server settings independent from Code Execution.
For Code Interpreter to work, all 3 conditions must be met:
  1. Code Interpreter enabled in this tab
  2. Code Interpreter capability set to Default On or Default Off in agent settings
  3. User manually toggles Code Interpreter in chat (when Default Off)
Missing any one disables Code Interpreter.