moai-adk 0.32.8__py3-none-any.whl → 0.41.0__py3-none-any.whl

moai_adk/templates/.claude/skills/moai-foundation-claude/reference/claude-code-cli-reference-official.md

@@ -0,0 +1,420 @@
+# Claude Code CLI Reference - Official Documentation Reference
+
+Source: https://code.claude.com/docs/en/cli-reference
+Updated: 2026-01-06
+
+## Overview
+
+The Claude Code CLI provides command-line access to Claude's capabilities with comprehensive options for customization, automation, and integration.
+
+## Basic Commands
+
+### Interactive Mode
+
+```bash
+claude
+```
+
+Starts Claude Code in interactive terminal mode.
+
+### Direct Query
+
+```bash
+claude "Your question or task"
+```
+
+Sends a single query and enters interactive mode.
+
+### Prompt Mode
+
+```bash
+claude -p "Your prompt"
+```
+
+Runs prompt, outputs response, and exits.
+
+### Continue Conversation
+
+```bash
+claude -c "Follow-up"
+```
+
+Continues the most recent conversation.
+
+### Resume Session
+
+```bash
+claude -r session_id "Continue task"
+```
+
+Resumes a specific session by ID.
+
+### Update CLI
+
+```bash
+claude update
+```
+
+Updates Claude Code to the latest version.
+
+## System Prompt Options
+
+### Replace System Prompt
+
+```bash
+claude -p "Task" --system-prompt "Custom instructions"
+```
+
+Warning: Removes Claude Code default capabilities.
+
+### Append to System Prompt
+
+```bash
+claude -p "Task" --append-system-prompt "Additional context"
+```
+
+Recommended: Preserves Claude Code functionality.
+
+### Load from File
+
+```bash
+claude -p "Task" --system-prompt-file prompt.txt
+```
+
+Loads system prompt from external file.
+
+## Tool Management
+
+### Specify Tools
+
+```bash
+claude -p "Task" --tools "Read,Write,Bash"
+```
+
+Explicitly lists available tools.
+
+### Allow Tools (Auto-approve)
+
+```bash
+claude -p "Task" --allowedTools "Read,Grep,Glob"
+```
+
+Auto-approves specified tools without prompts.
+
+### Tool Pattern Matching
+
+```bash
+claude -p "Task" --allowedTools "Bash(git:*)"
+```
+
+Allow specific command patterns only.
+
+### Multiple Patterns
+
+```bash
+claude -p "Task" --allowedTools "Bash(npm:*),Bash(git:*),Read"
+```
+
+### Disallow Tools
+
+```bash
+claude -p "Task" --disallowedTools "Bash,Write"
+```
+
+Prevents Claude from using specified tools.
+
+## Output Options
+
+### Output Format
+
+```bash
+claude -p "Task" --output-format text
+claude -p "Task" --output-format json
+claude -p "Task" --output-format stream-json
+```
+
+Available formats: text (default), json, stream-json
+
+### JSON Schema Validation
+
+```bash
+claude -p "Extract data" --json-schema '{"type": "object"}'
+```
+
+Validates output against JSON schema.
+
+### Schema from File
+
+```bash
+claude -p "Task" --json-schema-file schema.json
+```
+
+## Session Management
+
+### Fork Session
+
+```bash
+claude -p "Alternative approach" --fork-session session_id
+```
+
+Creates a new branch from existing session.
+
+### Maximum Turns
+
+```bash
+claude -p "Complex task" --max-turns 15
+```
+
+Limits conversation turns.
+
+## Agent Configuration
+
+### Use Specific Agent
+
+```bash
+claude -p "Review code" --agent code-reviewer
+```
+
+Uses defined sub-agent.
+
+### Dynamic Agent Definition
+
+```bash
+claude -p "Task" --agents '{
+  "my-agent": {
+    "description": "Agent purpose",
+    "prompt": "System prompt",
+    "tools": ["Read", "Grep"],
+    "model": "sonnet"
+  }
+}'
+```
+
+Defines agents inline via JSON.
+
+## Settings
+
+### Override Settings
+
+```bash
+claude -p "Task" --settings '{"model": "opus"}'
+```
+
+Overrides settings for this invocation.
+
+### Show Setting Sources
+
+```bash
+claude --setting-sources
+```
+
+Displays origin of each setting value.
+
+## Browser Integration
+
+### Enable Chrome
+
+```bash
+claude -p "Browse task" --chrome
+```
+
+Enables browser automation.
+
+### Disable Chrome
+
+```bash
+claude -p "Code task" --no-chrome
+```
+
+Disables browser features.
+
+## MCP Server Commands
+
+### Add MCP Server
+
+HTTP transport:
+```bash
+claude mcp add --transport http server-name https://url
+```
+
+Stdio transport:
+```bash
+claude mcp add --transport stdio server-name command args
+```
+
+SSE transport (deprecated):
+```bash
+claude mcp add --transport sse server-name https://url
+```
+
+### List MCP Servers
+
+```bash
+claude mcp list
+```
+
+### Get Server Details
+
+```bash
+claude mcp get server-name
+```
+
+### Remove MCP Server
+
+```bash
+claude mcp remove server-name
+```
+
+## Plugin Commands
+
+### Install Plugin
+
+```bash
+claude plugin install plugin-name
+claude plugin install owner/repo
+claude plugin install https://github.com/owner/repo.git
+claude plugin install plugin-name --scope project
+```
+
+### Uninstall Plugin
+
+```bash
+claude plugin uninstall plugin-name
+```
+
+### Enable/Disable Plugin
+
+```bash
+claude plugin enable plugin-name
+claude plugin disable plugin-name
+```
+
+### Update Plugin
+
+```bash
+claude plugin update plugin-name
+claude plugin update  # Update all
+```
+
+### List Plugins
+
+```bash
+claude plugin list
+```
+
+### Validate Plugin
+
+```bash
+claude plugin validate .
+```
+
+## Environment Variables
+
+### Configuration Variables
+
+- CLAUDE_API_KEY: API authentication key
+- CLAUDE_MODEL: Default model selection
+- CLAUDE_OUTPUT_FORMAT: Default output format
+- CLAUDE_TIMEOUT: Request timeout in seconds
+
+### Runtime Variables
+
+- CLAUDE_PROJECT_DIR: Current project directory
+- CLAUDE_CODE_REMOTE: Indicates remote execution
+- CLAUDE_ENV_FILE: Path to environment file
+
+### MCP Variables
+
+- MAX_MCP_OUTPUT_TOKENS: Maximum MCP output (default: 25000)
+- MCP_TIMEOUT: MCP server timeout in milliseconds
+
+### Update Control
+
+- DISABLE_AUTOUPDATER: Disable automatic updates
+
+## Exit Codes
+
+- 0: Success
+- 1: General error
+- 2: Permission denied or blocked operation
+
+## Complete Examples
+
+### CI/CD Code Review
+
+```bash
+claude -p "Review this PR for security issues" \
+  --allowedTools "Read,Grep,Glob" \
+  --append-system-prompt "Focus on OWASP Top 10 vulnerabilities" \
+  --output-format json \
+  --max-turns 5
+```
+
+### Automated Documentation
+
+```bash
+claude -p "Generate API documentation for src/" \
+  --allowedTools "Read,Glob,Write" \
+  --json-schema-file docs-schema.json
+```
+
+### Structured Data Extraction
+
+```bash
+claude -p "Extract all function signatures from codebase" \
+  --allowedTools "Read,Grep,Glob" \
+  --json-schema '{"type":"array","items":{"type":"object","properties":{"name":{"type":"string"},"params":{"type":"array"},"returns":{"type":"string"}}}}'
+```
+
+### Git Commit Message
+
+```bash
+git diff --staged | claude -p "Generate commit message" \
+  --allowedTools "Read" \
+  --output-format text
+```
+
+### Multi-Agent Workflow
+
+```bash
+claude -p "Analyze and refactor this module" \
+  --agents '{
+    "analyzer": {
+      "description": "Code analyzer",
+      "tools": ["Read", "Grep"],
+      "model": "haiku"
+    },
+    "refactorer": {
+      "description": "Code refactorer",
+      "tools": ["Read", "Write", "Edit"],
+      "model": "sonnet"
+    }
+  }'
+```
+
+## Best Practices
+
+### Security
+
+- Use --allowedTools to restrict capabilities
+- Avoid --dangerously-skip-permissions in untrusted environments
+- Validate input before passing to Claude
+
+### Performance
+
+- Use appropriate --max-turns for task complexity
+- Consider haiku model for simple tasks
+- Use --output-format json for programmatic parsing
+
+### Debugging
+
+- Use --setting-sources to troubleshoot configuration
+- Check exit codes for error handling
+- Use --output-format json for detailed response metadata
+
+### Automation
+
+- Always specify --allowedTools in scripts
+- Use --output-format json for reliable parsing
+- Handle errors with exit code checks
+- Log session IDs for debugging