### Install and Start MCP Server

Source: https://rysweet.github.io/amplihack/mcp_evaluation/USER_GUIDE

Example commands to install a generic MCP server using npm and start it, including a health check.

```bash
# Example: Installing a generic MCP server
# (Replace with your tool's actual installation)

# Install MCP server package
npm install -g @your-vendor/mcp-server

# Start the server
mcp-server start --port 3000

# Verify server is running
curl http://localhost:3000/health
# Should return: {"status": "healthy"}
```

--------------------------------

### Quick Start: Getting help

Source: https://rysweet.github.io/amplihack/claude/commands/ddd/prime

Command to access the complete guide and help.

```bash
/ddd:0-help

```

--------------------------------

### Project-Specific Settings Example

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Example of project-specific settings overriding plugin settings.

```json
{
  "amplihack": {
    "preferred_workflow": "INVESTIGATION_WORKFLOW",
    "custom_agents": ["./local-agents/domain-expert.md"],
    "disabled_hooks": ["PreCommit"]
  }
}

```

--------------------------------

### Installed Usage Example

Source: https://rysweet.github.io/amplihack/agent-bundle-generator-guide

Shows how to generate and run an agent bundle after cloning the repository.

```bash
# Generate bundle
amplihack bundle generate "your agent description" --output-dir ~/my-agent

# Run the bundle
cd ~/my-agent && uvx . agent-command

```

--------------------------------

### Install from Git

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Install directly from the repository.

```bash
pip install git+https://github.com/rysweet/amplihack.git
amplihack plugin install

```

--------------------------------

### Quick Start: Agent Generation Output

Source: https://rysweet.github.io/amplihack/GOAL_AGENT_GENERATOR_GUIDE

Example output from the amplihack new command, showing the agent generation process.

```text
Generating goal agent from: my_goal.md

[1/4] Analyzing goal prompt...
  Goal: Automated Security Audit
  Domain: security-analysis
  Complexity: moderate

[2/4] Creating execution plan...
  Phases: 4
  Estimated duration: 45 minutes

[3/4] Matching skills...
  Skills matched: 3
    - security-analyzer (85% match)
    - documenter (100% match)
    - generic-executor (60% match)

[4/4] Assembling agent bundle...
  Bundle name: security-automated-security-audit-agent

✓ Goal agent created successfully in 0.1s

Agent directory: ./goal_agents/security-automated-security-audit-agent

```

--------------------------------

### Manually Configure LSP

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Example JSON configuration for manually setting up LSP for Python projects.

```json
{
  "lsp": {
    "python": {
      "command": "pylsp",
      "enabled": true
    }
  }
}
```

--------------------------------

### Start Amplihack Tutorial

Source: https://rysweet.github.io/amplihack/tutorials/amplihack-tutorial

Command to invoke the guide agent and start the amplihack tutorial.

```bash
"Start the amplihack tutorial"
```

--------------------------------

### Quick Start: Write Your Goal

Source: https://rysweet.github.io/amplihack/GOAL_AGENT_GENERATOR_GUIDE

Example of creating a markdown file describing a goal for the agent generator.

```bash
cat > my_goal.md <<'EOF'
# Goal: Automated Security Audit

Scan Python code for security vulnerabilities and generate reports.

## Constraints
- Must complete within 30 minutes
- Should check for OWASP Top 10
- Must provide actionable recommendations

## Success Criteria
- Identifies at least 5 vulnerability types
- Generates prioritized report
- Includes code examples for fixes

## Context
This agent will help maintain secure codebases by automatically detecting common security issues.
EOF

```

--------------------------------

### Install from Marketplace

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Install from the Claude Code plugin marketplace.

```bash
claude plugin install github.com/rysweet/amplihack

```

--------------------------------

### Zero-Install Usage Example

Source: https://rysweet.github.io/amplihack/agent-bundle-generator-guide

Demonstrates how to generate and run an agent bundle directly from GitHub without installation.

```bash
# Generate an agent bundle
uvx --from git+https://github.com/rysweet/MicrosoftHackathon2025-AgenticCoding amplihack bundle generate \
  "Create an agent that monitors my system resources and alerts me when CPU or memory usage is high" \
  --output-dir ~/src/system-monitor

# Run your new agent
uvx --from ~/src/system-monitor system-monitor

```

--------------------------------

### Verify Installation

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Check that the plugin installed correctly.

```bash
amplihack plugin verify

```

--------------------------------

### Configuration 1: Manual PR Approval Setup

Source: https://rysweet.github.io/amplihack/features/power-steering/configuration

Example of setting up manual PR approval in USER_PREFERENCES.md.

```markdown
### 2026-01-23 10:00:00

**NEVER Merge PRs Without Explicit Permission**

NEVER merge PRs or commit directly to main without explicit user permission.
Always create PRs and wait for approval.


```

--------------------------------

### Direct Installation Example

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

Illustrates how to directly use the InteractiveInstaller to install a specific tool for a given platform.

```python
from amplihack.utils.prerequisites import InteractiveInstaller, Platform

installer = InteractiveInstaller(Platform.MACOS)
result = installer.install_tool("node")

if result.success:
    print("Node.js installed successfully!")
elif not result.user_approved:
    print("Installation declined by user")
else:
    print(f"Installation failed: {result.stderr}")

```

--------------------------------

### Basic Usage Example

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

Demonstrates how to use the PrerequisiteChecker to check and install dependencies interactively.

```python
from amplihack.utils.prerequisites import PrerequisiteChecker

checker = PrerequisiteChecker()
result = checker.check_and_install(interactive=True)

if result.all_available:
    print("All prerequisites installed!")
else:
    print(f"{len(result.missing_tools)} tools still missing")

```

--------------------------------

### Successful Installation Output

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

Example output demonstrating a successful installation process.

```text
======================================================================
MISSING PREREQUISITES DETECTED
======================================================================

Found 1 missing tool(s):
  - node

======================================================================

======================================================================
INSTALL NODE
======================================================================

The following command will be executed to install node:

  brew install node

This command may:
  - Require sudo password for system-level installation
  - Install dependencies automatically
  - Modify system packages or configuration

======================================================================

Do you want to proceed with installing node? [y/N]: y

Installing node...

[SUCCESS] node installed successfully

======================================================================
VERIFYING INSTALLATIONS
======================================================================

[SUCCESS] All prerequisites are now available!


```

--------------------------------

### Install GitHub CLI

Source: https://rysweet.github.io/amplihack/tutorials/platform-bridge-quickstart

Command to install the GitHub CLI using Homebrew.

```bash
brew install gh

```

--------------------------------

### Install Azure CLI

Source: https://rysweet.github.io/amplihack/tutorials/platform-bridge-quickstart

Command to install the Azure CLI using Homebrew.

```bash
brew install azure-cli

```

--------------------------------

### Install to custom path

Source: https://rysweet.github.io/amplihack/plugin/CLI_REFERENCE

Example of setting a custom path for plugin installation and then installing.

```bash
export AMPLIHACK_PLUGIN_PATH=~/custom/.claude/
amplihack plugin install

```

--------------------------------

### Example settings.json

Source: https://rysweet.github.io/amplihack/plugin/MIGRATION

An example of a minimal settings.json file containing project-specific overrides for agents, workflows, and preferred language.

```json
{
  "agents": {
    "custom_agents": ["./agents/custom/domain-expert.md", "./agents/custom/legacy-specialist.md"]
  },
  "workflows": {
    "default": "INVESTIGATION_WORKFLOW"
  },
  "preferred_language": "python"
}
```

--------------------------------

### Best Practice 1: Good Prompt Example

Source: https://rysweet.github.io/amplihack/remote-sessions/TUTORIAL

An example of a descriptive prompt for starting a remote session.

```bash
amplihack remote start "fix authentication token expiration bug in /api/auth/refresh endpoint"

```

--------------------------------

### Install via pip

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

The fastest way to install amplihack as a plugin.

```bash
pip install amplihack
amplihack plugin install

```

--------------------------------

### Troubleshooting: Install agent-framework-core

Source: https://rysweet.github.io/amplihack/SDK_ADAPTERS_GUIDE

Command to install the agent-framework-core package.

```bash
pip install agent-framework-core
```

--------------------------------

### Copilot SDK Installation

Source: https://rysweet.github.io/amplihack/SDK_ADAPTERS_GUIDE

Command to install the GitHub Copilot SDK.

```bash
pip install github-copilot-sdk

```

--------------------------------

### Windows Installation Commands

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

Example commands for installing various tools on Windows using winget and powershell.

```json
{
    "node": ["winget", "install", "OpenJS.NodeJS"],
    "npm": ["winget", "install", "OpenJS.NodeJS"],
    "uv": ["powershell", "-c", "irm https://astral.sh/uv/install.ps1 | iex"],
    "git": ["winget", "install", "Git.Git"],
    "claude": ["npm", "install", "-g", "@anthropic-ai/claude-code"]
}

```

--------------------------------

### Installation Behavior - Interactive

Source: https://rysweet.github.io/amplihack/reference/amplifier-command

Example of the interactive installation prompt for Amplifier.

```bash
Installing Amplifier from git+https://github.com/microsoft/amplifier...
This will install Amplifier as a uv tool.
Continue? [y/N] y
✓ Amplifier CLI installed

```

--------------------------------

### Check your setup

Source: https://rysweet.github.io/amplihack/tutorials/first-docs-site

Verify that Python and Git are installed and configured correctly.

```bash
python --version   # Should be 3.8+
git --version      # Any recent version

```

--------------------------------

### Troubleshooting: Verify agent-framework-core installation

Source: https://rysweet.github.io/amplihack/SDK_ADAPTERS_GUIDE

Command to verify the installation of agent-framework-core.

```python
python -c "from agent_framework import Agent; print('OK')"
```

--------------------------------

### Install plugin for CI/CD

Source: https://rysweet.github.io/amplihack/plugin/MIGRATION

Example of how to install the Amplihack plugin in a GitHub Actions workflow for CI/CD pipelines.

```yaml
# .github/workflows/test.yml
- name: Install amplihack plugin
  run: |
    pip install amplihack
    amplihack plugin install

```

--------------------------------

### Scenario 1: First-time setup with Anthropic

Source: https://rysweet.github.io/amplihack/document_driven_development/phases/05_testing_and_verification

Example output from the initial setup wizard when configuring Anthropic as the provider.

```bash
uvx --from git+https://...@next amplifier
amplifier

```

```text
Welcome to Amplifier!
First time? Let's get you set up.
Provider? [1] Anthropic [2] OpenAI [3] Azure OpenAI [4] Ollama: 1 API key: •••••••• Get one: https://console.anthropic.com/settings/keys ✓ Saved to ~/.amplifier/keys.env
Model? [1] claude-sonnet-4-5 [2] claude-opus-4 [3] custom: 1 ✓ Using claude-sonnet-4-5
Profile? [1] dev [2] base [3] full: 1 ✓ Activated profile: dev
Ready! Starting chat...

```

--------------------------------

### Claude CLI Utilities Example

Source: https://rysweet.github.io/amplihack/DEVELOPING_AMPLIHACK

Example demonstrating how to get the Claude CLI path and install it if not found.

```python
from amplihack.utils.claude_cli import (
    get_claude_cli_path,
    install_claude_cli
)

claude_path = get_claude_cli_path(auto_install=False)
if not claude_path:
    if install_claude_cli():
        claude_path = get_claude_cli_path(auto_install=False)


```

--------------------------------

### Direct Execution Mode

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example output indicating the recipe runner is starting in direct mode.

```text
[dev-orchestrator] starting recipe runner (direct mode)...

```

--------------------------------

### Example Usage: Starting a New Feature

Source: https://rysweet.github.io/amplihack/claude/commands/ddd/0-help

This example demonstrates the sequence of commands to initiate and complete a new feature using the DDD workflow.

```bash
# Load context (optional but recommended)
/ddd:prime

# Phase 1: Plan the feature
/ddd:1-plan Add user authentication with JWT tokens

# Phase 2: Update all docs
/ddd:2-docs

# Review the changes, iterate if needed
# When satisfied, commit the docs yourself

# Phase 3: Plan code implementation
/ddd:3-code-plan

# Review the code plan, approve to continue

# Phase 4: Implement and test
/ddd:4-code

# Test, provide feedback, iterate until working

# Phase 5: Finalize
/ddd:5-finish

# Cleanup, push, PR (with your explicit approval at each step)


```

--------------------------------

### Rule 3: Ruthless Simplicity - Say More with Less - Good Example

Source: https://rysweet.github.io/amplihack/DOCUMENTATION_GUIDELINES

A simplified and actionable example demonstrating clear instructions for authentication setup.

```bash
export AUTH_TOKEN="your-token"

```

--------------------------------

### Install and Run the Generated Agent

Source: https://rysweet.github.io/amplihack/AGENT_MEMORY_QUICKSTART

Commands to navigate to the agent's directory, install dependencies, and run the agent.

```bash
cd goal_agents/incident-memory-agent
python -m pip install -r requirements.txt
python main.py

```

--------------------------------

### Configuration 3: Relaxed Workflow Setup

Source: https://rysweet.github.io/amplihack/features/power-steering/configuration

Example of modifying considerations.yaml to change blockers to warnings.

```yaml
# Change all blockers to warnings
- id: tests_written
  severity: warning # Changed from blocker

- id: ci_status
  severity: warning # Changed from blocker

# Keep only critical blockers
- id: zero_bs_compliance
  severity: blocker # Keep as blocker


```

--------------------------------

### Run the recipe

Source: https://rysweet.github.io/amplihack-recipe-runner/quickstart.html

Execute the 'hello.yaml' recipe using the recipe-runner-rs binary.

```bash
recipe-runner-rs hello.yaml
```

--------------------------------

### Local Preview Server

Source: https://rysweet.github.io/amplihack/tutorials/first-docs-site

Start a local server to preview the generated documentation site.

```bash
mkdocs serve

```

--------------------------------

### Configuration 4: Custom Team Requirements Setup

Source: https://rysweet.github.io/amplihack/features/power-steering/configuration

Example of adding custom team requirements to considerations.yaml.

```yaml
- id: security_review
  category: Code Quality & Philosophy Compliance
  question: Has the security team reviewed this change for sensitive code?
  description: Ensures security team sign-off for authentication, authorization, or data handling changes
  severity: blocker
  checker: generic # Uses keyword matching
  enabled: true

- id: compliance_checklist
  category: Session Completion & Progress
  question: Has the compliance checklist been completed for this feature?
  description: Validates regulatory compliance requirements documented
  severity: warning
  checker: generic
  enabled: true


```

--------------------------------

### Create a file called hello.yaml

Source: https://rysweet.github.io/amplihack-recipe-runner/quickstart.html

Create a basic YAML recipe file named 'hello.yaml' with a simple greeting.

```yaml
name: "hello-world"
description: "A minimal recipe to verify your setup"
version: "1.0.0"
context:
  greeting: "Hello from the Recipe Runner!"
steps:
  - id: "greet"
    command: "echo '{{greeting}}'"
```

--------------------------------

### Configuration 2: Strict Local Testing Setup

Source: https://rysweet.github.io/amplihack/features/power-steering/configuration

Example of setting up strict local testing in USER_PREFERENCES.md.

```markdown
### 2026-01-23 10:00:00

**Step 13 Local Testing - NO EXCEPTIONS**

EVERY PR MUST have local testing completed. Step 13 in DEFAULT_WORKFLOW.md
is MANDATORY and cannot be skipped through rationalization.

**Implementation Requirements:**

- MUST execute at least 2 test scenarios (1 simple + 1 complex) locally
- MUST document test results in PR description
- MUST verify no regressions before committing


```

--------------------------------

### Run the deployment script

Source: https://rysweet.github.io/amplihack/tutorials/first-docs-site

Command to execute the deployment script.

```bash
python deploy_docs.py
```

--------------------------------

### Parallel Workstreams Command

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example of a /dev command that can be split into parallel workstreams.

```bash
/dev build a REST API and a React webui for user management

```

--------------------------------

### Clone the repository

Source: https://rysweet.github.io/amplihack-recipe-runner/quickstart.html

Clone the amplihack-recipe-runner repository, build it in release mode, and optionally copy the binary to your PATH.

```bash
git clone https://github.com/rysweet/amplihack-recipe-runner.git
cd amplihack-recipe-runner

# Build in release mode
cargo build --release

# The binary is at target/release/recipe-runner-rs
# Optionally copy it to your PATH:
cp target/release/recipe-runner-rs ~/.local/bin/
```

--------------------------------

### First /dev Command

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example of the first /dev command to fix a bug.

```bash
/dev fix the login timeout bug

```

--------------------------------

### Quick Start Setup

Source: https://rysweet.github.io/amplihack/WORKSPACE_PATTERN

Steps to clone the amplihack workspace, add a project as a git submodule, set up project context with AGENTS.md, and start working.

```bash
# 1. Create your amplihack workspace
git clone https://github.com/rysweet/MicrosoftHackathon2025-AgenticCoding.git my-workspace
cd my-workspace

# 2. Add your project (existing or new)
git submodule add <your-project-git-url> my-project
# OR: git submodule add git@github.com:yourname/your-project.git my-project

# 3. Set up project context
cd my-project
cat > AGENTS.md << 'EOF'
# My Project Context

This file provides guidance to AI agents working with this project.

## Project Overview
[Brief description of what this project does]

## Working in This Project

When working on this project:
- All project files belong in this directory
- Use `ai_working/` for temporary files
- Reference files with `@my-project/` prefix
- Follow our design principles at @my-project/docs/DESIGN.md

## Key Technologies
- [List your main technologies/frameworks]

## Development Workflow
- [Your specific workflow notes]
EOF

# 4. Start working
cd ..
uvx claude

```

--------------------------------

### Execution Complete Output Example

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example of the final summary output after the Dev Orchestrator completes its execution, detailing the task, type, workstreams, and a summary of accomplishments.

```text
# Dev Orchestrator -- Execution Complete

**Task**: [your task]
**Type**: [Q&A | Development | Investigation | Operations]
**Workstreams**: [number]

## Summary

[PR links, goal status, what was accomplished, any remaining work]
```

--------------------------------

### Simplified Updates Example (After Migration)

Source: https://rysweet.github.io/amplihack/plugin/MIGRATION

Example showing time taken for updates after migration, with a single command updating all projects via the plugin.

```text
amplihack plugin install --upgrade  # 30s
# All projects updated instantly
Total: 30 seconds (3× faster)
```

--------------------------------

### Running Recipes with Different Environments

Source: https://rysweet.github.io/amplihack/recipes/cli-examples

Command-line examples for running the 'smart-deploy' recipe with different context configurations for staging and production deployments.

```bash
# Deploy to staging (coverage check applies)
amplihack recipe run smart-deploy \
  --context '{"environment": "staging"}'

# Deploy to prod (both checks apply)
amplihack recipe run smart-deploy \
  --context '{"environment": "production"}'

```

--------------------------------

### Docker-Specific Setup for Amplihack

Source: https://rysweet.github.io/amplihack/troubleshooting/memory-consent-issues

A Dockerfile example that installs Amplihack and sets AMPLIHACK_MEMORY_AUTO_ACCEPT to true for non-interactive container environments.

```dockerfile
FROM python:3.11
RUN pip install amplihack

# Non-interactive by default in containers
ENV AMPLIHACK_MEMORY_AUTO_ACCEPT=true

CMD ["amplihack"]

```

--------------------------------

### Troubleshooting Tutorial Start

Source: https://rysweet.github.io/amplihack/tutorials/amplihack-tutorial

Commands to troubleshoot issues with the tutorial not starting, including checking the version and explicit agent invocation.

```bash
# Ensure you're in amplihack environment
amplihack --version

# Try explicit agent invocation
Task(subagent_type="guide", prompt="Start tutorial from Section 1")
```

--------------------------------

### Investigation + Implementation Command

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example of a /dev command for tasks requiring investigation before implementation.

```bash
/dev investigate how the payment service handles retries, then add exponential backoff

```

--------------------------------

### Force Single Workstream

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Example of setting an environment variable to force single-session execution.

```bash
export AMPLIHACK_MAX_DEPTH=0
/dev build a REST API and a React webui
# Falls back to single-session execution

```

--------------------------------

### Get Code Context for Memory Example

Source: https://rysweet.github.io/amplihack/blarify_quickstart

Python code to query code context for a given memory ID.

```python
from amplihack.memory.neo4j import Neo4jConnector, BlarifyIntegration

with Neo4jConnector() as conn:
    integration = BlarifyIntegration(conn)

    # Get code context for a memory
    context = integration.query_code_context("memory-id-here")

    for func in context["functions"]:
        print(f"{func['name']} in {func['file_path']}")
```

--------------------------------

### Execution Round Example

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Demonstrates the iterative process of the Dev Orchestrator where a builder agent attempts to achieve a goal, reflects on the outcome, and starts a new round if necessary.

```text
Round 1: Builder implements fix
Round 1 reflection: GOAL_STATUS: PARTIAL -- tests not updated
Round 2: Builder adds missing tests
Round 2 reflection: GOAL_STATUS: ACHIEVED
```

--------------------------------

### Extract Text from Scanned PDFs (Setup)

Source: https://rysweet.github.io/amplihack/claude/skills/pdf/SKILL

Shows the necessary pip installations for extracting text from scanned PDFs using pytesseract and pdf2image.

```bash
# Requires: pip install pytesseract pdf2image

```

--------------------------------

### Common Patterns Examples

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Provides examples of common command patterns for various development tasks, including bug fixes, new features, investigations, parallel tasks, and code reviews.

```bash
# Bug fix
/dev fix the 500 error on the /users endpoint

# New feature
/dev add OAuth login with Google

# Investigation only
/dev investigate why database queries are slow on the dashboard

# Parallel features
/dev add rate limiting and add request logging to the API

# Investigation then implement
/dev understand the existing test structure then add tests for the auth module

# Code review
/dev review PR #42 for security issues

# Simple Q&A (no workflow overhead)
/dev what does the circuit breaker pattern do?
```

--------------------------------

### Prerequisites and Execution

Source: https://rysweet.github.io/amplihack/GOAL_AGENT_GENERATOR_GUIDE

Steps to set up and run a generated agent, including installation, API key configuration, and the command to start execution.

```bash
cd <agent-directory>
python main.py

```

--------------------------------

### Manual Override Example

Source: https://rysweet.github.io/amplihack/tutorials/dev-orchestrator-tutorial

Shows how to re-run the Dev Orchestrator with more specific instructions for single-round execution, including explicit success criteria.

```bash
/dev fix the login timeout bug -- ensure existing tests pass and add a regression test
```

--------------------------------

### Quick Start

Source: https://rysweet.github.io/amplihack/BENCHMARKING

Steps to clone eval-recipes, copy agent configs, set API key, and run the benchmark.

```bash
# 1. Clone eval-recipes
git clone https://github.com/microsoft/eval-recipes.git ~/eval-recipes
cd ~/eval-recipes

# 2. Copy our agent configs
cp -r /path/to/amplihack/.claude/agents/eval-recipes/* data/agents/ 

# 3. Set API key
echo "ANTHROPIC_API_KEY=sk-ant-...". > .env

# 4. Run benchmark
uv run scripts/run_benchmarks.py \
  --agent-filter "name=amplihack" \
  --task-filter "name=linkedin_drafting" \
  --num-trials 1

```

--------------------------------

### Example Audit Entry

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

An example of a successful installation entry in the audit log.

```json
{
  "timestamp": "2025-01-18T10:30:00Z",
  "tool": "node",
  "platform": "macos",
  "command": ["brew", "install", "node"],
  "user_approved": true,
  "success": true,
  "exit_code": 0,
  "error_message": null
}

```

--------------------------------

### Install azlin via uvx from GitHub

Source: https://rysweet.github.io/amplihack/remote-sessions/TUTORIAL

Installs the azlin tool using uvx from a GitHub repository and displays its help message.

```bash
# Install via uvx from GitHub (not available on PyPI)
uvx --from git+https://github.com/rysweet/azlin --python 3.11 azlin --help
```

--------------------------------

### ClaudeLauncher prepare_launch() Example

Source: https://rysweet.github.io/amplihack/DEVELOPING_AMPLIHACK

Example demonstrating how to use the prepare_launch() method of the ClaudeLauncher.

```python
launcher = ClaudeLauncher()
if launcher.prepare_launch():
    # Ready to launch
    pass
```

--------------------------------

### Simplified Updates Example (Before Migration)

Source: https://rysweet.github.io/amplihack/plugin/MIGRATION

Example showing time taken for updates before migration, requiring individual project updates.

```text
cd project1 && amplihack upgrade  # 30s
cd project2 && amplihack upgrade  # 30s
cd project3 && amplihack upgrade  # 30s
Total: 90 seconds
```

--------------------------------

### Quick Start

Source: https://rysweet.github.io/amplihack/PROFILE_MANAGEMENT

Set profile via environment variable and install with profile filtering.

```bash
# Set profile via environment variable
export AMPLIHACK_PROFILE=amplihack://profiles/coding

# Install with profile filtering
amplihack install
# Result: Only 9/32 agents copied (72% reduction)

# Or launch with profile filtering
amplihack claude
# Result: Only 9/32 agents staged to working directory

```

--------------------------------

### Quick Workflow Example

Source: https://rysweet.github.io/amplihack/commands/COMMAND_SELECTION_GUIDE

An example of a common development workflow involving fixing errors, building a module, and checking compliance.

```bash
/fix import                    # Fix import errors
/modular-build auth-module     # Build new module
/analyze .                     # Check philosophy compliance

```

--------------------------------

### Declined Installation Output

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

Example output showing a declined installation process.

```text
======================================================================
INSTALL NPM
======================================================================

The following command will be executed to install npm:

  brew install node

This command may:
  - Require sudo password for system-level installation
  - Install dependencies automatically
  - Modify system packages or configuration

======================================================================

Do you want to proceed with installing npm? [y/N]: n

[SKIPPED] npm installation declined by user


```

--------------------------------

### Running a Recipe

Source: https://rysweet.github.io/amplihack-recipe-runner/examples-tutorials.html

Example command to run a specific recipe tutorial.

```bash
recipe-runner-rs examples/tutorials/01-hello-world.yaml
```

--------------------------------

### Declined Installation Entry

Source: https://rysweet.github.io/amplihack/INTERACTIVE_INSTALLATION

An example of a declined installation entry in the audit log.

```json
{
  "timestamp": "2025-01-18T10:31:00Z",
  "tool": "npm",
  "platform": "macos",
  "command": ["brew", "install", "node"],
  "user_approved": false,
  "success": false,
  "exit_code": -1,
  "error_message": "User declined installation"
}

```

--------------------------------

### Troubleshooting: Plugin directory not found - Re-run installation

Source: https://rysweet.github.io/amplihack/plugin/INSTALLATION

Re-run installation command.

```bash
amplihack plugin install

```

--------------------------------

### Start Command Examples

Source: https://rysweet.github.io/amplihack/remote-sessions/CLI_REFERENCE

Demonstrates various ways to use the `amplihack remote start` command with different options and prompts.

```bash
# Single task
amplihack remote start "implement user authentication"

# Multiple tasks
amplihack remote start "task one" "task two" "task three"

# With custom options
amplihack remote start --size l --max-turns 20 "complex refactoring"

# Use ultrathink mode (deep multi-agent analysis)
amplihack remote start --command ultrathink "analyze architecture"

# Long-running task with higher turn limit
amplihack remote start --max-turns 30 "comprehensive refactoring"

# Specify region (useful for quota management)
amplihack remote start --region eastus "my task"

```

--------------------------------

### Install plugin

Source: https://rysweet.github.io/amplihack/plugin/MIGRATION

Installs the centralized amplihack plugin.

```bash
# Install plugin
amplihack plugin install
```

--------------------------------

### Migration Guide Example

Source: https://rysweet.github.io/amplihack/document_driven_development/core_concepts/retcon_writing

A placeholder for a migration guide, if necessary.

```markdown
# Migration Guide: v1.0 → v2.0


```

--------------------------------

### Agent steps

Source: https://rysweet.github.io/amplihack-recipe-runner/quickstart.html

Example of a recipe using an agent step to invoke an AI agent.

```yaml
name: "analyze-project"
description: "Analyze a codebase with an AI agent"
version: "1.0.0"
context:
  repo_path: "."
steps:
  - id: "analyze"
    agent: "amplihack:core:architect"
    prompt: "Analyze the project at {{repo_path}} and summarize its structure"
    output: "analysis"
    parse_json: true

  - id: "report"
    command: "echo 'Analysis complete'"
    condition: "analysis"
```

--------------------------------

### Manual Hook Installation Example

Source: https://rysweet.github.io/amplihack/COPILOT_CLI

Example of manually installing a specific Git hook using symbolic links and setting execute permissions.

```bash
# Install specific hook
ln -s ../../.github/hooks/pre-commit .git/hooks/pre-commit
chmod +x .git/hooks/pre-commit

```

--------------------------------

### Create Feature Branch

Source: https://rysweet.github.io/amplihack/tutorials/platform-bridge-quickstart

Creates a new Git branch for the tutorial.

```bash
git checkout -b tutorial/platform-bridge-test

```

--------------------------------

### Mini SDK Example

Source: https://rysweet.github.io/amplihack/tutorials/GOAL_SEEKING_AGENT_TUTORIAL

Command to generate an agent using the Mini SDK.

```bash
amplihack new --file goal.md --sdk mini

```

--------------------------------

### Claude Agent SDK Installation

Source: https://rysweet.github.io/amplihack/SDK_ADAPTERS_GUIDE

Command to install the claude-agent-sdk.

```bash
pip install claude-agent-sdk

```

--------------------------------

### Running the tutorial script

Source: https://rysweet.github.io/amplihack/tutorials/learning-agent-refactor-tutorial

Command to execute the Python tutorial script.

```bash
uv run python tutorial_learning_agent.py

```

--------------------------------

### Install amplihack

Source: https://rysweet.github.io/amplihack/HOOK_CONFIGURATION_GUIDE

Command to install amplihack from a Git repository.

```bash
# Install amplihack
uvx --from git+https://github.com/rysweet/MicrosoftHackathon2025-AgenticCoding amplihack install

# Hooks are installed to ~/.claude/settings.json
# Everything works!

```

--------------------------------

### Run Site Generation

Source: https://rysweet.github.io/amplihack/tutorials/first-docs-site

Execute the Python script to generate the documentation site.

```bash
python generate_docs.py

```

--------------------------------

### Development Workflow Examples

Source: https://rysweet.github.io/amplihack/recipes/quick-reference

Examples for full feature implementation, fast bug fixes, and code investigation using specific recipes.

```bash
# Full feature implementation
amplihack recipe run default-workflow \
  --context task_description="Add JWT auth" \
  --context repo_path="."

# Fast bug fix
amplihack recipe run quick-fix \
  --context task_description="Fix null pointer in UserService"

# Code investigation
amplihack recipe run investigation \
  --context task_description="How does auth work?" \
  --context focus_area="src/auth/"

```

--------------------------------

### Install Python dependencies

Source: https://rysweet.github.io/amplihack/blarify_quickstart

Command to install Python dependencies, assuming they are already installed with amplihack.

```bash
# Already installed with amplihack
pip install -e .
```