gemini-system

サブエージェント経由を推奨。 Gemini出力は大きくなりがちなため。

| 状況 | 方法 |

|------|------|

| コードベース分析 | サブエージェント経由（推奨） |

| ライブラリ調査 | サブエージェント経由（推奨） |

| マルチモーダル | サブエージェント経由（推奨） |

| 短い質問 (1-2文回答) | 直接呼び出しOK |

| Task | Gemini | Codex |

|------|--------|-------|

| リポジトリ全体理解 | ✓ | |

| ライブラリ調査 | ✓ | |

| マルチモーダル (PDF/動画/音声) | ✓ | |

| 最新ドキュメント検索 | ✓ | |

| 設計判断 | | ✓ |

| デバッグ | | ✓ |

| コード実装 | | ✓ |

| Situation | Trigger Examples |

|-----------|------------------|

| Research | 「調べて」「リサーチ」 / "Research" "Investigate" |

| Library docs | 「ライブラリ」「ドキュメント」 / "Library" "Docs" |

| Codebase analysis | 「コードベース全体」 / "Entire codebase" |

| Multimodal | 「PDF」「動画」「音声」 / "PDF" "Video" "Audio" |

• Design decisions (use Codex)
• Debugging (use Codex)
• Code implementation (use Codex)
• Simple file operations (do directly)

Recommended: Subagent Pattern

Use Task tool with subagent_type='general-purpose' to preserve main context.

```

Task tool parameters:

• subagent_type: "general-purpose"
• run_in_background: true (optional, for parallel work)
• prompt: |

Research: {topic}

gemini -p "{research question}" 2>/dev/null

Save full output to: .claude/docs/research/{topic}.md

Return CONCISE summary (5-7 bullet points).

```

Direct Call (Short Questions Only)

For quick questions expecting brief answers:

```bash

gemini -p "Brief question" 2>/dev/null

```

CLI Options Reference

```bash

# Codebase analysis

gemini -p "{question}" --include-directories . 2>/dev/null

# Multimodal (PDF/video/audio)

gemini -p "{prompt}" < /path/to/file.pdf 2>/dev/null

# JSON output

gemini -p "{question}" --output-format json 2>/dev/null

```

Workflow (Subagent)

1. Spawn subagent with Gemini research prompt
2. Continue your work → Subagent runs in parallel
3. Receive summary → Subagent returns key findings
4. Full output saved → .claude/docs/research/{topic}.md

1. Ask Gemini in English
2. Receive response in English
3. Synthesize and apply findings
4. Report to user in Japanese

Save Gemini research results to:

```

.claude/docs/research/{topic}.md

```

This allows Claude and Codex to reference the research later.

Pre-Implementation Research

```bash

gemini -p "Research best practices for {feature} in Python 2025.

Include:

• Common patterns and anti-patterns
• Library recommendations (with comparison)
• Performance considerations
• Security concerns
• Code examples" 2>/dev/null

```

Repository Analysis

```bash

gemini -p "Analyze this repository:

1. Architecture overview
2. Key modules and responsibilities
3. Data flow between components
4. Entry points and extension points
5. Existing patterns to follow" --include-directories . 2>/dev/null

```

Library Research

See: references/lib-research-task.md

Multimodal Analysis

```bash

# Video

gemini -p "Analyze video: main concepts, key points, timestamps" < tutorial.mp4 2>/dev/null

# PDF

gemini -p "Extract: API specs, examples, constraints" < api-docs.pdf 2>/dev/null

# Audio

gemini -p "Transcribe and summarize: decisions, action items" < meeting.mp3 2>/dev/null

```

| Workflow | Steps |

|----------|-------|

| New feature | Gemini research → Codex design review |

| Library choice | Gemini comparison → Codex decision |

| Bug investigation | Gemini codebase search → Codex debug |

• 1M token context: Entire repositories at once
• Google Search: Latest information and docs
• Multimodal: Native PDF/video/audio processing
• Fast exploration: Quick overview before deep work
• Shared context: Results saved for Claude/Codex