メニュー
Guide for customizing OhMyCode's AI personality and behavior via system prompt. Use when user wants to change how the AI behaves, its tone, or its instructions.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Run OhMyCode benchmarks — score any provider/model with token tracking. Use when user wants to benchmark, evaluate, test performance, or compare models.
Generate high-quality tests for OhMyCode modules. Use when user wants to create, add, or generate tests for a module or file.
Run tests and analyze results for OhMyCode. Use when user wants to run, check, or verify tests — or after any code change.
Guide for debugging OhMyCode issues. Use when user reports errors, unexpected behavior, or connection problems.
Guide for adding a new feature to OhMyCode. Use when user wants to add functionality that goes beyond existing extension points (tools/providers). Always start by reading docs/DEVELOPMENT_GUIDE.md.
Guide for adding a new LLM provider to OhMyCode. Use when user wants to connect a new AI model backend.
| name | customize-system-prompt |
| description | Guide for customizing OhMyCode's AI personality and behavior via system prompt. Use when user wants to change how the AI behaves, its tone, or its instructions. |
Change the AI assistant's personality, behavior rules, or domain expertise without writing code.
Create a OHMYCODE.md (or CLAUDE.md) file in your project root:
# Project Instructions
You are working on a Django REST API project.
## Rules
- Always use type hints
- Prefer class-based views over function-based
- Write docstrings for all public functions
- Use Black formatting style
## Tech Stack
- Python 3.12, Django 5.0, PostgreSQL
- pytest for testing, factory_boy for fixtures
OhMyCode auto-detects this file and injects it into the system prompt.
Scope: Only applies when running ohmycode in this directory.
Add to ~/.ohmycode/config.json:
{
"system_prompt_append": "Always respond in Chinese. Prefer simple solutions over clever ones. Never use global variables."
}
Scope: Applies to ALL conversations.
Edit ohmycode/core/system_prompt.py → build_system_prompt() to change the base role description or add new sections.
Scope: Permanent change to the codebase.
# OHMYCODE.md
You are a patient coding tutor. Explain concepts step by step.
When the user makes a mistake, explain why it's wrong before showing the fix.
Use analogies to explain complex concepts.
# OHMYCODE.md
You are a strict senior engineer doing code review.
Point out every issue: bugs, performance, readability, security.
Don't write code for the user — give them guidance to fix it themselves.
Rate code quality on a 1-10 scale.
# OHMYCODE.md
You are an expert in machine learning with PyTorch.
When suggesting solutions, prefer PyTorch over TensorFlow.
Always consider GPU memory usage and training efficiency.
Reference relevant papers when explaining algorithms.
ohmycode/core/system_prompt.py → build_system_prompt() assembles the prompt in this order:
OHMYCODE.md / CLAUDE.md (found via find_project_instructions())~/.ohmycode/memory/MEMORY.mdconfig.system_prompt_appendOHMYCODE.md focused — don't repeat what's already in the base promptohmycode -p "Describe yourself" --mode autoOHMYCODE.md is searched upward from current directory — put it at the repo root