From claude-commands
Clarifies which Gemini models can combine code_execution with JSON mode/structured outputs. Essential for building Gemini-powered apps needing both code execution and controlled generation.
How this skill is triggered — by the user, by Claude, or both
Slash command
/claude-commands:gemini-code-execution-json-modeThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
**CAN combine code_execution with JSON mode/structured outputs.**
CAN combine code_execution with JSON mode/structured outputs.
From Gemini 3 Developer Guide:
"Gemini 3 allows you to combine Structured Outputs with built-in tools, including Grounding with Google Search, URL Context, and Code Execution."
CANNOT combine code_execution with JSON mode/controlled generation.
Error returned:
INVALID_ARGUMENT: Unable to submit request because controlled generation
is not supported with Code Execution tool.
| Model | Code Execution | JSON Mode | Both Together |
|---|---|---|---|
| gemini-3-pro-preview | YES | YES | YES |
| gemini-2.0-flash | YES | YES | NO |
| gemini-2.5-flash | NO | YES | N/A |
For Your Project dice rolling:
tool_requests array# Gemini 3: Can enable code_execution with JSON mode
if model_name in GEMINI_3_MODELS:
enable_code_execution = True # Safe with JSON mode
# Gemini 2.x: Must disable code_execution when using JSON mode
if model_name in ["gemini-2.0-flash", "gemini-2.5-flash"]:
enable_code_execution = False # Would cause INVALID_ARGUMENT
2025-12-16
npx claudepluginhub jleechanorg/claude-commands --plugin claude-commandsProvides usage notes for Gemini 3 API via google-genai SDK, covering ThinkingConfig parameters, code execution tool, structured JSON output, and integration checklist.
Invokes Google Gemini CLI for complex reasoning, research, and AI tasks in headless mode. Supports preview models, fallbacks, and session continuation.
Runs Google's Gemini CLI for code generation, review, analysis, web research via Google Search, and codebase architecture investigation. Use for second AI opinions, real-time web data, or parallel code tasks.