### Manual Waza Installation Script Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/getting-started.mdx

Manually install the Waza binary by downloading it from the releases page and placing it in your system's PATH. Ensure you replace `<version>` and the platform suffix with your specific details.

```bash
# Example: macOS ARM64 (Apple Silicon) — replace <version> with the latest tag (e.g. v0.34.0)
curl -fSL -o waza https://github.com/microsoft/waza/releases/download/<version>/waza-darwin-arm64
chmod +x waza
sudo mv waza /usr/local/bin/waza
```

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

### Serve with custom port

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of starting the Waza server on a custom port.

```bash
waza serve --port 8080
```

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

### waza init command examples

Source: https://github.com/microsoft/waza/blob/main/skills/waza/SKILL.md

Examples of how to use the `waza init` command.

```bash
# Initialize in current directory
waza init

# Initialize in a named directory
waza init my-eval-suite
```

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

### Real-World Scenario Example

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Illustrates a practical decision-making scenario based on model performance and cost. This output is for informational purposes and guides production choices.

```bash
echo "💼 Real-world use case:"
echo ""
echo "Scenario: Choosing a model for your production skill"
echo ""
echo "Results Summary:"
echo "  Claude Sonnet:  Score 0.85 | Cost: ~$0.01/task"
echo "  Claude Opus:    Score 0.92 | Cost: ~$0.05/task"
echo "  GPT-4o:         Score 0.90 | Cost: ~$0.03/task"
echo ""
echo "Decision:"
echo "  ✅ Use Claude Sonnet in prod (best cost/quality ratio)"
echo "  🔄 Monitor Claude Opus quarterly (check for improvements)"
echo "  ❌ Skip GPT-4o for now (not worth 5x cost for 5% quality gain)"
```

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

### Install Waza CLI

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/index.mdx

Installs the Waza CLI tool on macOS, Linux, or Windows Git Bash/MSYS/Cygwin. Run this command to get started.

```bash
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
waza init my-eval-suite
cd my-eval-suite
waza new skill my-skill
waza run my-skill -v
waza serve  # View results in the dashboard
```

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

### Serve with TCP for JSON-RPC

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of starting the Waza server with TCP for JSON-RPC communication.

```bash
waza serve --tcp :9000
```

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

### Install Project Dependencies

Source: https://github.com/microsoft/waza/blob/main/site/README.md

Installs all the necessary packages listed in your project's package.json.

```bash
npm install
```

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

### Examples

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Various examples of how to use the `waza run` command with different options.

```bash
# Run all tasks
waza run eval.yaml -v
```

```bash
# Run specific skill
waza run code-explainer
```

```bash
# Specify fixtures directory
waza run eval.yaml -c ./fixtures -v
```

```bash
# Save results
waza run eval.yaml -o results.json
```

```bash
# Filter to specific tasks
waza run eval.yaml --task "basic*" --task "edge*"
```

```bash
# Multiple models (parallel)
waza run eval.yaml --model gpt-4o --model claude-sonnet-4.6
```

```bash
# Use a different judge model for LLM-as-judge graders
waza run eval.yaml --model gpt-4o --judge-model claude-opus-4.6
```

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

### Copy Waza Example Directory

Source: https://github.com/microsoft/waza/blob/main/examples/README.md

Copy an existing waza example directory to a new location for modification. This is the first step in using an example as a template.

```bash
cp -r examples/grader-showcase my-eval
cd my-eval
```

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

### Install Waza Extension

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Installs the Waza extension for use with the Azure Developer CLI.

```bash
# Install the extension
azd ext install microsoft.azd.waza
```

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

### Installation

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Install waza using the provided script for macOS/Linux or Windows Git Bash/MSYS/Cygwin, and verify the installation.

```bash
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
waza --version
```

```powershell
irm https://raw.githubusercontent.com/microsoft/waza/main/install.ps1 | iex
waza --version
```

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

### Install waza CLI

Source: https://github.com/microsoft/waza/blob/main/docs/INTEGRATION-TESTING.md

Installs the waza command-line interface. Run this before configuring or running tests.

```bash
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
```

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

### Install Waza on Windows (PowerShell)

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/releases.mdx

Use this PowerShell command to install Waza on Windows. This command downloads, verifies, and installs the Windows binary.

```powershell
irm https://raw.githubusercontent.com/microsoft/waza/main/install.ps1 | iex
```

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

### Show Example Directory Structure

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Displays the directory structure of the code explainer example, limited to a depth of 2.

```bash
# Navigate to project root
cd /path/to/waza

# Show the example structure
echo "📁 Example directory structure:"
tree examples/code-explainer/ -L 2
```

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

### Waza Tokens Configuration Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of configuring per-file token limits in `.waza.yaml`.

```yaml
tokens:
  limits:
    defaults:
      "*.md": 2000
      "skills/**/SKILL.md": 5000
    overrides:
      "skills/complex-skill/SKILL.md": 7500
```

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

### waza generate command example

Source: https://github.com/microsoft/waza/blob/main/skills/waza/SKILL.md

Example of how to use the `waza generate` command.

```bash
# Generate eval from SKILL.md
waza generate path/to/SKILL.md
```

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

### waza grade examples

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Examples of using the waza grade command.

```bash
# Grade all tasks from a previous run
waza grade eval.yaml --results results.json

# Grade a specific task
waza grade eval.yaml --results results.json --task "basic-function"

# Use a different judge model
waza grade eval.yaml --results results.json --judge-model claude-opus-4.6

# Save graded results for comparison
waza grade eval.yaml --results results.json -o graded.json
```

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

### Install Waza

Source: https://github.com/microsoft/waza/blob/main/DEMO-SCRIPT.md

Installs the waza framework using a curl script. Ensure you are in a clean environment before running.

```bash
# Clean environment
cd ~/demo && rm -rf waza-demo
mkdir waza-demo && cd waza-demo

# Install waza (macOS/Linux or Windows Git Bash/MSYS/Cygwin)
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
```

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

### Run All Grader Showcase Examples

Source: https://github.com/microsoft/waza/blob/main/examples/grader-showcase/README.md

Execute all available grader examples in the showcase using the waza run command.

```bash
waza run examples/grader-showcase/eval.yaml -v
```

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

### Quick Start: Initialize Project

Source: https://github.com/microsoft/waza/blob/main/README.md

Initialize a new Waza project and navigate into it.

```bash
# Initialize a new project
waza init my-project && cd my-project
```

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

### .waza.yaml Configuration Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of a root '.waza.yaml' configuration file for project-level settings.

```yaml
# Root of waza project
paths:
  skills: skills
  evals: evals
  results: results

# Generated/discovered eval and task naming
files:
  evalFile: eval.yaml
  taskGlob: tasks/*.yaml
  taskFileSuffix: .yaml

# Model defaults
defaults:
  model: claude-sonnet-4.6
  timeout: 300
  workers: 0

# Cache settings
cache:
  enabled: false
  dir: .waza-cache

# Token budget configuration
tokens:
  warningThreshold: 2500
  fallbackLimit: 1000
  limits:
    defaults:
      "SKILL.md": 500
      "references/**/*.md": 1000
      "*.md": 2000
    overrides:
      "README.md": 3000

# Cloud storage (optional)
storage:
  provider: azure-blob
  accountName: "myteamwaza"
  containerName: "waza-results"
  enabled: true
```

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

### waza run command examples

Source: https://github.com/microsoft/waza/blob/main/skills/waza/SKILL.md

Examples of how to use the `waza run` command with various flags.

```bash
# Run with default mock engine
waza run path/to/eval.yaml --context-dir path/to/fixtures

# Verbose output with results saved
waza run eval.yaml -c ./fixtures -v -o results.json

# Filter to specific tasks
waza run eval.yaml -t "task-name-1" -t "task-name-2"

# Parallel execution
waza run eval.yaml --parallel --workers 4

# Save per-task transcripts
waza run eval.yaml --transcript-dir ./transcripts
```

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

### Development Setup and Build Commands

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/about.mdx

Instructions for setting up the Waza development environment, including cloning, building, testing, and linting the project.

```bash
# Clone repository
git clone https://github.com/microsoft/waza.git
cd waza

# Set up Go environment
go version  # Requires 1.26+

# Build
make build

# Test
make test

# Lint
make lint

# Run
./waza --help
```

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

### Waza Suggest Command Examples

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Examples of using `waza suggest` for previewing, applying, and JSON output.

```bash
# Preview suggestions
waza suggest skills/code-explainer --dry-run

# Write eval/task/fixture files
waza suggest skills/code-explainer --apply

# JSON output
waza suggest skills/code-explainer --format json
```

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

### Waza CLI Commands via azd

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Examples of common Waza commands when installed as an Azure Developer CLI extension.

```bash
azd waza init my-project
azd waza run eval.yaml
azd waza serve
```

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

### Full Azure Storage Setup and Workflow

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/azure-storage.mdx

Demonstrates the complete workflow for setting up Azure Storage with Waza, running evaluations, and managing results. This includes Azure CLI commands for setup and Waza commands for configuration and execution.

```bash
# 1. Set up Azure (one-time)
az login
az group create -n waza-eval -l eastus
az storage account create -n myteamwaza -g waza-eval --sku Standard_LRS
az storage container create -n waza-results -a myteamwaza

# 2. Configure project (one-time)
cat > .waza.yaml <<EOF
storage:
  provider: azure-blob
  accountName: "myteamwaza"
  containerName: "waza-results"
  enabled: true
EOF

# 3. Run evaluations (repeatable)
waza run evals/code-explainer/eval.yaml -v -o results-$(date +%s).json

# 4. Browse team's results
waza results list --limit 20

# 5. Compare your run with previous run
waza results compare 20250226-143025-abc123 20250226-135410-def456
```

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

### Quality evaluation example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of running the `waza quality` command for a specific skill path.

```bash
waza quality skills/code-explainer
```

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

### Filter Tasks During Evaluation

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Run an evaluation and filter the tasks to be executed based on specific patterns. This example includes tasks starting with 'explain-' and 'validate-'.

```bash
./waza-bin run eval.yaml --task "explain-*" --task "validate-*" -v
```

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

### Build Waza Binary and Get Help

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Build the Waza binary using 'make build' and then display help information for the 'run' command.

```bash
make build
./waza-bin run examples/code-explainer/eval.yaml --help
```

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

### waza dev workflow example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of using the waza dev command for iterative improvement.

```bash
waza dev code-explainer --target high --auto
```

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

### Quick Start .waza.yaml Configuration

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/waza-yaml.mdx

Sets up default paths for skills, evaluations, and results, along with file naming conventions and execution parameters like model and timeout.

```yaml
# .waza.yaml
paths:
  skills: skills/
  evals: evals/
  results: results/

files:
  evalFile: eval.yaml
  taskGlob: tasks/*.yaml
  taskFileSuffix: .yaml

defaults:
  model: claude-sonnet-4.6
  timeout: 300
  workers: 0

```

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

### Initialize Waza Project and Create Skill

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/quick-start.mdx

Creates a new Waza project directory and then generates a new skill within that project. Shows the file structure created.

```bash
mkdir my-eval-suite
cd my-eval-suite
waza init
waza new skill my-skill
```

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

### Example with Missing Required Skills

Source: https://github.com/microsoft/waza/blob/main/examples/required-skills-demo/README.md

This configuration has missing skills in `skill_directories` that are listed in `required_skills`. A clear error is reported before the evaluation starts, listing the missing skills and search paths.

```yaml
config:
  skill_directories:
    - ./skills/azure-deploy  # Only has azure-deploy
  required_skills:
    - azure-deploy
    - azure-prepare
    - azure-validate
```

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

### Install Waza Binary

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Installs waza using a script. Ensure you are on macOS/Linux or using Windows Git Bash/MSYS/Cygwin.

```bash
# Option A: Binary install (macOS/Linux or Windows Git Bash/MSYS/Cygwin)
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
```

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

### Initialize Waza Project

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Creates a new Waza project directory and initializes its structure, including a default skill.

```bash
mkdir my-eval-suite
cd my-eval-suite
waza init
```

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

### GitHub Actions example using azd extension

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/ci-cd.mdx

This snippet shows how to set up the Azure Developer CLI and install the waza extension in a GitHub Actions workflow.

```yaml
name: Evaluate Skills

on:
  push:
    branches: [main, develop]
  pull_request:
    branches: [main]
    paths:
      - 'evals/**'
      - 'skills/**'
      - '.github/workflows/eval.yml'

permissions: 
  contents: read
  pull-requests: write

jobs:
  evaluate:
    runs-on: ubuntu-latest
    strategy:
      matrix:
        model: [gpt-4o, claude-sonnet-4.6, claude-opus-4]
      max-parallel: 3
    timeout-minutes: 30
    
    steps:
      - uses: actions/checkout@v4
      
      - name: Install waza
        run: |
          curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
          echo "$HOME/bin" >> $GITHUB_PATH
      
      - name: Run evaluations (${{ matrix.model }})
        run: |
          waza run \
            --model "${{ matrix.model }}" \
            --output "results-${{ matrix.model }}.json" \
            --verbose
      
      - name: Upload results
        uses: actions/upload-artifact@v4
        with:
          name: results-${{ matrix.model }}
          path: results-${{ matrix.model }}.json
          retention-days: 30
      
      - name: Comment PR with results
        if: github.event_name == 'pull_request'
        uses: actions/github-script@v7
        with:
          script: |
            const fs = require('fs');
            const results = JSON.parse(fs.readFileSync('results-${{ matrix.model }}.json', 'utf8'));
            github.rest.issues.createComment({
              issue_number: context.issue.number,
              owner: context.repo.owner,
              repo: context.repo.repo,
              body: `### Eval Results: ${{ matrix.model }}\n\n✅ **Passed:** ${results.summary.succeeded}\n❌ **Failed:** ${results.summary.failed}`
            });

```

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

### Customize GitHub Actions Workflow for Waza

Source: https://github.com/microsoft/waza/blob/main/docs/SKILLS_CI_INTEGRATION.md

An example of a GitHub Actions workflow file configured to install Waza and run skill evaluations on pull requests or pushes to the main branch. It includes uploading evaluation results as an artifact.

```yaml
on:
  pull_request:
    branches: [ main ]
    paths:
      - 'SKILL.md'
      - 'eval/**'
  push:
    branches: [ main ]

jobs:
  evaluate-skill:
    runs-on: ubuntu-latest
    steps:
      - uses: actions/checkout@v4
      
      - name: Install Waza
        run: curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
      
      - name: Run Evaluation
        run: waza run eval/eval.yaml --verbose --output results.json
      
      - name: Upload Results
        if: always()
        uses: actions/upload-artifact@v4
        with:
          name: evaluation-results
          path: results.json
```

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

### Quick Start: Create Skill

Source: https://github.com/microsoft/waza/blob/main/README.md

Create a new skill scaffold.

```bash
# Create a new skill
waza new skill my-skill
```

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

### Install waza in GitHub Actions

Source: https://github.com/microsoft/waza/blob/main/docs/CI-CD-GUIDE.md

Installs the waza CLI tool in a GitHub Actions environment. Ensure the waza binary is in your PATH.

```yaml
name: Install waza
run: curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
```

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

### Start Live Evaluation with Waza CLI

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/dashboard-explore.mdx

Start a Waza server to serve results and then run an evaluation with the --live flag to stream progress to the dashboard.

```bash
# Start a live evaluation and watch it in the dashboard
waza serve --results-dir ./results &
waza run eval.yaml --context-dir ./fixtures --live
```

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

### Install Waza from Source

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/getting-started.mdx

Installs Waza by building from source code. Requires Go 1.26+ and Git LFS. Clones the repository, initializes Git LFS, and builds the binary.

```bash
git clone https://github.com/microsoft/waza.git
cd waza
git lfs install
git lfs pull
go build -o waza ./cmd/waza
./waza --version
```

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

### Install Waza via Bash Script

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/quick-start.mdx

Installs Waza using a curl script. Verifies the installation by checking the version.

```bash
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
waza --version
```

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

### Inputs Prompt File Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of using 'prompt_file' for specifying prompt text.

```yaml
inputs:
  prompt_file: prompts/review-instructions.md
```

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

### Install Waza using Azure Developer CLI

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/quick-start.mdx

Adds the Waza extension to the Azure Developer CLI and verifies its installation.

```bash
azd ext source add -n waza -t url -l https://raw.githubusercontent.com/microsoft/waza/main/registry.json
azd ext install microsoft.azd.waza
azd waza --version
```

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

### Typical Waza Development Workflow

Source: https://github.com/microsoft/waza/blob/main/docs/GETTING-STARTED.md

A step-by-step guide for the common development cycle in Waza, from initializing a project and creating a skill to defining its metadata, writing evaluation tasks, adding fixtures, running local evaluations, and committing changes.

```bash
# 1. Start project
waza init my-project && cd my-project
```

```bash
# 2. Create a skill
waza new skill my-skill
```

```bash
# 3. Define the skill
edit skills/my-skill/SKILL.md
```

```bash
# 4. Write evaluation tasks
edit evals/my-skill/tasks/*.yaml
```

```bash
# 5. Add test fixtures
cp ~/my-fixtures/* evals/my-skill/fixtures/
```

```bash
# 6. Run evaluations locally
waza run my-skill -v
```

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

### Install waza via binary

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/ci-cd.mdx

Shell commands to download and install the waza binary on Linux/macOS.

```bash
# Linux / macOS
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
export PATH="$HOME/bin:$PATH"
```

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

### Verify Waza Installation

Source: https://github.com/microsoft/waza/blob/main/docs/SKILLS_CI_INTEGRATION.md

Checks if Waza has been installed correctly by displaying its version. This command is used after installing Waza via binary or source.

```bash
waza --version
```

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

### Initialize a Waza Project

Source: https://github.com/microsoft/waza/blob/main/DEMO-SCRIPT.md

Sets up the directory structure for a new waza project, including folders for skills, evals, and CI/CD configurations.

```bash
waza init ./my-skill-project
```

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

### Example with All Required Skills Present

Source: https://github.com/microsoft/waza/blob/main/examples/required-skills-demo/README.md

This configuration includes all necessary skills in `skill_directories` and `required_skills`. Validation passes, and the evaluation proceeds normally.

```yaml
config:
  skill_directories:
    - ./skills/azure-deploy
    - ./skills/azure-prepare
    - ./skills/azure-validate
  required_skills:
    - azure-deploy
    - azure-prepare
    - azure-validate
```

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

### Install Waza using Bash script

Source: https://github.com/microsoft/waza/blob/main/README.md

Download and install the latest pre-built binary with the Bash install script on macOS, Linux, or Windows Bash environments.

```bash
curl -fsSL https://raw.githubusercontent.com/microsoft/waza/main/install.sh | bash
```

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

### Initialize New Waza Project

Source: https://github.com/microsoft/waza/blob/main/DEMO-SCRIPT.md

Initializes a new Waza project in the specified directory.

```bash
# Initialize new project
waza init my-project
```

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

### Weighted scoring example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/graders.mdx

Example demonstrating how to use weights for graders.

```yaml
graders:
  - type: text
    name: critical_check
    weight: 3.0 # Counts 3×
    config:
      regex_match: ["deployed"]

  - type: text
    name: nice_to_have
    weight: 0.5 # Counts 0.5×
    config:
      contains: [summary]

  - type: code
    name: basic_length
    # weight omitted → defaults to 1.0
    config:
      assertions:
        - "len(output) > 50"
```

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

### waza cache clear examples

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Examples for clearing the cache.

```bash
# Clear default cache
waza cache clear

# Clear custom cache directory
waza cache clear --cache-dir .my-cache
```

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

### Install Waza from Source

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Builds Waza from its source code. Requires Go 1.26+ and Git LFS.

```bash
git clone https://github.com/microsoft/waza.git
cd waza
git lfs install
git lfs pull
go build -o waza ./cmd/waza
```

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

### Initialize Waza Project and Create Skill

Source: https://github.com/microsoft/waza/blob/main/docs/TUTORIAL.md

Initializes a Waza project directory structure and then creates a new skill within it.

```bash
# Create a waza project with directory structure
waza init my-awesome-skill

# Then create skills within it
waza new skill my-skill
```

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

### JavaScript Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/graders.mdx

An example of JavaScript assertions for checking agent output.

```yaml
- type: code
  name: js_checks
  config:
    language: javascript
    assertions:
      - "output.length > 50"
      - "output.includes('hello')"
```

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

### Significance Testing Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/statistical-fields.mdx

An example demonstrating non-significant results.

```json
{
  "task": "deploy-app",
  "model_a": { "score": 0.80, "trials": 5 },
  "model_b": { "score": 0.82, "trials": 5 },
  "difference": 0.02,
  "is_significant": false,
  "p_value": 0.18
}
```

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

### Display Project Structure

Source: https://github.com/microsoft/waza/blob/main/DEMO-SCRIPT.md

Shows the directory structure of the code explainer example project.

```bash
tree examples/code-explainer
```

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

### Install waza from source

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/ci-cd.mdx

Commands to clone the waza repository and build the binary if Go is available.

```bash
git clone https://github.com/microsoft/waza.git
cd waza
go build -o waza ./cmd/waza
```

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

### Build Waza from Source

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Builds waza from source code. Navigate to the waza directory before running.

```bash
# Option B: Build from source
cd /path/to/waza
make build
```

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

### Verify Waza Installation

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Checks if waza is installed correctly by displaying its help message.

```bash
# Verify installation
waza run --help
```

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

### Real-World Scenario Example

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Illustrates a real-world scenario for deploying an Azure function, detailing the main skill, utility skills, and error handling skills involved in the orchestration.

```bash
echo "🌍 Real-world scenario:"
echo "  Task: 'Deploy Azure function'"
echo "  Required skills:"
echo "    1. azure-functions (main)"
echo "    2. azure-cli (utility)"
echo "    3. error-handler (error recovery)"
echo ""
echo "  Waza ensures all skills are available, invokes them correctly,"
echo "  and validates the full orchestration end-to-end."
```

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

### CSV Format Example

Source: https://github.com/microsoft/waza/blob/main/README.md

Example of the expected format for a CSV dataset.

```csv
prompt,expected_output,language
"Explain this function","Function explanation",python
"Review this code","Code review",javascript
```

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

### Initialize New Project with GitHub Actions

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/about.mdx

Create a new Waza project and automatically set up a CI/CD workflow using GitHub Actions.

```bash
waza init my-project
# Creates .github/workflows/eval.yml
```

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

### Update Notification Example

Source: https://github.com/microsoft/waza/blob/main/README.md

Example of an update notification message from Waza.

```text
A newer version of waza is available: v0.24.0 → v0.28.0. Run: waza update
```

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

### Not authenticated error example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of the error message shown when not authenticated.

```bash
Error: not authenticated — run "copilot login" first
```

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

### Initialize Evaluation Suite Interactively

Source: https://github.com/microsoft/waza/blob/main/docs/SKILLS_CI_INTEGRATION.md

Starts an interactive wizard to help configure a new evaluation suite for a skill. Navigate to your skill directory before running.

```bash
# Navigate to your skill directory
cd your-skill

# Run the interactive wizard
waza init eval

# Follow the prompts to configure your evaluation
```

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

### Start Local Development Server

Source: https://github.com/microsoft/waza/blob/main/site/README.md

Starts a local development server for previewing your site during development. The server typically runs at localhost:4321.

```bash
npm run dev
```

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

### Task Tags Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'tags' field for a task.

```yaml
tags:
  - basic
  - happy-path
  - python
```

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

### Quick Start: Run Evaluations

Source: https://github.com/microsoft/waza/blob/main/README.md

Run evaluations for a skill.

```bash
# Run evaluations
waza run my-skill
```

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

### Initialize Waza Project

Source: https://github.com/microsoft/waza/blob/main/docs/GETTING-STARTED.md

Use this command to create a new Waza project directory and initialize the workspace. It sets up the necessary directories for skills, evaluations, and CI/CD.

```bash
mkdir my-skills-repo
cd my-skills-repo
waza init
```

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

### Task Description Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'description' field for a task.

```yaml
description: "Test that the agent explains a simple Python function correctly"
```

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

### Show Example CI Workflow

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Displays a snippet of a GitHub Actions CI workflow file that demonstrates how Waza can be integrated into automated pipelines.

```bash
echo "🔄 Example CI workflow:"
cat examples/ci/README.md | head -60
```

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

### Task Name Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'name' field for a task.

```yaml
name: "Basic Usage - Python Function"
```

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

### Task ID Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'id' field for a task.

```yaml
id: basic-usage-001
```

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

### timeout_seconds Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of 'timeout_seconds' within the 'config' section.

```yaml
config:
  timeout_seconds: 300 # 5 minutes
```

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

### Initialize Waza Project

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Initialize a new Waza project with a standard directory structure. The --no-skill flag skips the initial skill creation prompt.

```bash
waza init my-skills --no-skill
```

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

### trials_per_task Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of 'trials_per_task' within the 'config' section.

```yaml
config:
  trials_per_task: 3
```

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

### version Field Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'version' field in eval.yaml.

```yaml
version: "1.0"
```

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

### skill Field Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'skill' field in eval.yaml.

```yaml
skill: code-explainer
```

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

### Launch the Waza Dashboard

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/dashboard-explore.mdx

Start the waza dashboard server to visualize evaluation results. Specify the directory containing your results.

```bash
waza serve --results-dir ./results
```

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

### description Field Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'description' field in eval.yaml.

```yaml
description: "Evaluates agent's ability to explain Python code"
```

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

### name Field Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of the 'name' field in eval.yaml.

```yaml
name: code-explainer-eval
```

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

### Evaluation Spec Format Example

Source: https://github.com/microsoft/waza/blob/main/skills/waza/SKILL.md

Example of the YAML format for defining an evaluation.

```yaml
name: my-eval
skill: my-skill
version: "1.0"
executor: mock          # or copilot-sdk
tasks:
  - id: task-1
    name: "Describe the task"
    prompt: "Your prompt to the agent"
    expected: "Expected behavior"
    validators:
      - type: code
        config:
          language: go
      - type: text
        config:
          pattern: "expected pattern"
```

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

### Create Demo Eval Directory Structure

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Sets up the necessary directory structure for a Waza prompt grading demonstration. This includes 'tasks' and 'fixtures' subdirectories.

```bash
mkdir -p waza-prompt-demo/{tasks,fixtures}
```

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

### Quick Start: Check Skill Readiness

Source: https://github.com/microsoft/waza/blob/main/README.md

Check if a skill is ready for submission.

```bash
# Check skill readiness
waza check my-skill
```

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

### Running the Custom Agent Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/custom-agents.mdx

Command to run the custom agent evaluation and its expected output, showing task progression and results.

```bash
cd examples/custom-agent
waza run eval.yaml -v

# Output:
# Running: security-reviewer eval
# Task: injection-detection ... ✓ PASS
# Task: crypto-review ... ✓ PASS
# Task: data-handling ... ⚠ PARTIAL (4/5 validators)
# Task: infrastructure-scan ... ✓ PASS
#
# Results saved to results.json
```

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

### Create New Starlight Project

Source: https://github.com/microsoft/waza/blob/main/site/README.md

Use this command to initialize a new Astro project with the Starlight template.

```bash
npm create astro@latest -- --template starlight
```

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

### Start the interactive dashboard

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Command to start the Waza interactive dashboard.

```bash
waza serve
```

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

### List all models in table format

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/cli.mdx

Example of listing all models in a table format.

```bash
# List all models in table format
waza models
```

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

### Verify Waza Installation (Source)

Source: https://github.com/microsoft/waza/blob/main/docs/GUIDE.md

Checks if the Waza binary built from source is installed and accessible.

```bash
./waza --version
```

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

### Define Python Sample File

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Creates a Python file named 'sample.py' within the demo fixtures directory. This file contains a simple function to be explained in the prompt grading task.

```python
def normalize_price(cents):
    return cents / 100.0
```

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

### Run the example security-reviewer agent

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/guides/custom-agents.mdx

This command runs a pre-built example of a custom security-reviewer agent using Waza.

```bash
waza run examples/custom-agent/eval.yaml -v
```

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

### Check Waza Version and Help

Source: https://github.com/microsoft/waza/blob/main/DEMO-SCRIPT.md

Displays the installed waza version and general help information. Useful for verifying installation and understanding available commands.

```bash
waza --version
waza --help
```

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

### Confidence Intervals Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/statistical-fields.mdx

An example of how confidence intervals are represented in the results JSON.

```json
{
  "task": "code-review",
  "trials": 10,
  "mean_score": 0.82,
  "bootstrap_ci": {
    "lower_bound": 0.75,
    "upper_bound": 0.88,
    "confidence_level": 0.95
  }
}
```

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

### Example Profile Output

Source: https://github.com/microsoft/waza/blob/main/README.md

Example of the output format for the 'waza tokens profile' command.

```text
📊 my-skill: 1,722 tokens (detailed ✓), 8 sections, 4 code blocks
   ⚠️  no workflow steps detected
```

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

### Display Example Task Grader Configuration

Source: https://github.com/microsoft/waza/blob/main/docs/DEMO-GUIDE.md

Shows the beginning of the code-task.yaml file to illustrate task-specific grader configurations.

```bash
# Show one task file to explain task-specific graders
echo "📄 Example task with graders:"
head -40 examples/grader-showcase/tasks/code-task.yaml
```

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

### Complete Token Limits Configuration Example

Source: https://github.com/microsoft/waza/blob/main/docs/TOKEN-LIMITS.md

A comprehensive JSON configuration for token limits, demonstrating defaults with various glob patterns and specific overrides.

```json
{
  "description": "Token limits for skill repository",
  "defaults": {
    "SKILL.md": 5000,
    "references/*.md": 2000,
    "references/**/*.md": 2000,
    "references/test-templates/*.md": 1500,
    "*.md": 4000
  },
  "overrides": {
    "README.md": 4000
  }
}
```

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

### Outcomes Example

Source: https://github.com/microsoft/waza/blob/main/site/src/content/docs/reference/schema.mdx

Example of 'outcomes' rule, specifying expected execution outcomes.

```yaml
expected:
  outcomes:
    - type: task_completed
    - type: tool_called
      tool_name: code_analyzer
```

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

### Waza Workspace Auto-Detection Examples

Source: https://github.com/microsoft/waza/blob/main/docs/GETTING-STARTED.md

Demonstrates how Waza automatically detects the project structure (project mode, standalone mode, multi-skill) and adapts commands accordingly. Examples cover running from different directories within a project or a standalone skill.

```bash
cd my-skills-repo
waza run              # Runs all skills' evals
```

```bash
waza run code-explainer  # Run one skill
```

```bash
cd my-skills-repo/skills/code-explainer
waza check            # Finds SKILL.md in parent, checks readiness
```

```bash
cd my-translator
waza run evals/eval.yaml  # Self-contained
```

```bash
waza check               # Checks current skill
```

```bash
cd sibling-skills
# Scans ./*/SKILL.md and ./evals/*/eval.yaml
waza run              # Run all discovered skills
```