Real-world examples for every scenario.
# Review all your open PRs with default settings
warden
# Preview what would be fixed (no changes)
warden --dry-run
# Security-focused review with full test suite
warden --reviewers security,testing --test-strategy full
# Fast review with affected tests only
warden --review-depth standard --test-strategy affected# Review your open PRs with defaults (standard depth, affected tests)
warden
# Review specific author's PRs
warden --author octocat
# Review closed PRs
warden --state closed --limit 5
# Review all PRs (open + closed)
warden --state all# Dry run - preview what would be fixed without making changes
warden --dry-run
# Dry run with verbose output
warden --dry-run --verbose
# Dry run and save report
warden --dry-run --save-report preview-report.md# Quick review with 1 generalist reviewer
warden --review-depth standard
# Same as:
warden --reviewers generalist# Security + Performance specialists
warden --review-depth thorough
# Same as:
warden --reviewers security,performance# Full coverage: Security + Performance + Architecture
warden --review-depth comprehensive
# Same as:
warden --reviewers security,performance,architecture# Security-focused review only
warden --reviewers security,testing --review-focus security
# Performance-critical code review
warden --reviewers performance,architecture
# Maintainability and testing focus
warden --reviewers maintainability,testing
# Maximum coverage (5 reviewers)
warden --reviewers security,performance,architecture,maintainability,testing# Documentation PR - no tests needed
warden --test-strategy none
# Config changes - auto-skip tests for certain file types
warden --skip-tests-for .yml,.yaml,.json,.md,.txt# Test only changed packages (fastest)
warden --test-strategy affected
# Test affected packages with custom timeout
warden --test-strategy affected --test-timeout 600# Run entire test suite (thorough)
warden --test-strategy full
# Full tests with longer timeout
warden --test-strategy full --test-timeout 900# Affected packages + dependencies
warden --test-strategy smart
# Smart testing with pre-validation
warden --test-strategy smart --test-before-fix# Only test when fixing Critical issues
warden --test-on-severity critical --test-strategy affected
# Only test for Critical + High
warden --test-on-severity critical,high --test-strategy smart
# Custom test command
warden --test-command "npm run test:ci"# High-confidence fixes only (90%+)
warden --fix-strategy conservative
# Conservative with full test suite
warden --fix-strategy conservative --test-strategy full
# Conservative for production code
warden \
--fix-strategy conservative \
--test-strategy full \
--reviewers security,testing# Fix most issues (70%+ confidence)
warden --fix-strategy balanced
# Balanced with affected tests
warden --fix-strategy balanced --test-strategy affected# Attempt all fixes including complex refactors
warden --fix-strategy aggressive
# Aggressive with limits
warden --fix-strategy aggressive --max-fixes-per-tier 20
# Aggressive for draft PR (preview first)
warden --fix-strategy aggressive --dry-run# Preview before committing
warden --auto-commit-on-success false
# Limit fixes per tier
warden --max-fixes-per-tier 10
# Require review before push (see SAFETY.md)
warden --require-review-before-push# Post findings as PR comment
warden --comment-on-pr
# Update existing comment instead of adding new
warden --comment-on-pr --update-comment
# Use custom comment template
warden --comment-on-pr --comment-template .github/warden-comment.md# Add labels based on findings
warden --auto-label
# Custom label prefix
warden --auto-label --label-prefix bot-
# Labels: security, performance, architecture, needs-tests, etc.# Analyze many PRs efficiently
warden --limit 50 --max-parallel-prs 20
# Limit total subagents for resource constraints
warden --max-parallel-agents 15
# Balance speed and resources
warden --limit 10 --max-parallel-prs 5 --max-parallel-agents 10# Reuse workspace for speed (same repo)
warden --reuse-workspace
# Keep workspace for debugging
warden --keep-workspace
# Custom workspace location
warden --workspace-dir /custom/workspace# Cache PR data for 1 hour
warden --cache-pr-data 3600
# Cache analysis results
warden --cache-analysis
# Custom cache location
warden --cache-location /tmp/warden-cache# JSON output
warden --output-format json --save-report warden-report.json
# Markdown report
warden --output-format markdown --save-report warden-report.md
# HTML report
warden --output-format html --save-report warden-report.html
# Text to console (default)
warden --output-format text# Verbose logging
warden --verbose
# Quiet mode (minimal output)
warden --quiet
# Quiet with report file
warden --quiet --save-report report.md
# Only show high severity and above
warden --severity high# Save metrics over time
warden --save-metrics .warden/metrics.json
# Combine with reporting
warden \
--save-report warden-$(date +%Y%m%d).md \
--save-metrics .warden/metrics.json# Ignore generated code
warden --ignore-paths "vendor/,*.pb.go,*_generated.go"
# Ignore third-party code
warden --ignore-paths "third_party/,external/,deps/"
# Multiple patterns
warden --ignore-paths "vendor/,node_modules/,dist/,build/,*.min.js"# Only review source code
warden --focus-paths "src/,cmd/"
# Only review specific component
warden --focus-paths "auth/,security/"
# Combine focus and ignore
warden --focus-paths "src/" --ignore-paths "src/generated/"# Skip large files (> 5000 lines)
warden --max-file-size 5000
# Skip generated files automatically
warden --skip-generated
# Disable generated file skipping
warden --skip-generated false# Send summary to Slack
warden --notify-slack https://hooks.slack.com/services/YOUR/WEBHOOK
# Combine with commenting
warden \
--comment-on-pr \
--notify-slack $SLACK_WEBHOOK# Update Jira ticket
warden --update-jira PROJ-123
# Update multiple tickets
warden --update-jira PROJ-123,PROJ-124# POST summary JSON to webhook
warden --webhook https://api.example.com/pr-review
# Custom integration
warden \
--webhook https://internal.example.com/warden \
--output-format json# Go project with gofmt and golangci-lint
warden \
--language go \
--formatter "gofmt -s -w" \
--linter "golangci-lint run" \
--test-command "go test -v ./..."
# Focus on Go source only
warden --language go --ignore-paths "*.pb.go,*_generated.go,vendor/"# Python with black and pytest
warden \
--language python \
--formatter "black --line-length 100" \
--linter "ruff check" \
--test-command "pytest -v"
# Python with type checking
warden \
--language python \
--test-command "pytest && mypy ."# Node.js with prettier and eslint
warden \
--language javascript \
--formatter "prettier --write" \
--linter "eslint --fix" \
--test-command "npm test"
# TypeScript with type checking
warden \
--language typescript \
--test-command "npm test && tsc --noEmit"# Rust with cargo fmt and clippy
warden \
--language rust \
--formatter "cargo fmt" \
--linter "cargo clippy -- -D warnings" \
--test-command "cargo test"warden \
--reviewers security,testing \
--review-focus security \
--test-strategy full \
--fix-strategy conservative \
--comment-on-pr \
--auto-label \
--notify-slack $SLACK_WEBHOOK \
--save-report security-audit-$(date +%Y%m%d).mdwarden \
--state all \
--test-strategy affected \
--test-on-severity critical \
--max-fixes-per-tier 5 \
--quiet \
--save-report draft-fixes.mdwarden \
--reviewers security,performance,architecture \
--test-strategy smart \
--fix-strategy conservative \
--comment-on-pr \
--update-comment \
--auto-label \
--save-report infrastructure-audit.md \
--save-metrics .warden/metrics.json \
--verbosewarden \
--limit 100 \
--max-parallel-prs 25 \
--max-parallel-agents 50 \
--test-strategy affected \
--test-on-severity critical,high \
--fix-strategy balanced \
--reuse-workspace \
--cache-pr-data 3600 \
--quiet \
--save-report batch-$(date +%Y%m%d-%H%M%S).json \
--output-format jsonwarden \
--test-strategy none \
--reviewers maintainability \
--fix-strategy aggressive \
--ignore-paths "src/,cmd/,lib/" \
--focus-paths "docs/,README.md,*.md"warden \
--reviewers security,performance,architecture,testing \
--test-strategy full \
--fix-strategy conservative \
--test-timeout 1200 \
--comment-on-pr \
--auto-label \
--notify-slack $SLACK_WEBHOOK \
--webhook https://api.internal.com/release-gate \
--save-report release-$(date +%Y%m%d).md \
--save-metrics .warden/release-metrics.json \
--verbose# GitHub Actions / GitLab CI
warden \
--dry-run \
--reviewers security,performance \
--test-strategy affected \
--comment-on-pr \
--update-comment \
--auto-label \
--output-format markdown \
--save-report ci-report.md# Review only files changed since last Warden run
warden \
--incremental-review \
--test-strategy affected \
--fix-strategy balanced# Use custom review rules file
warden --review-rules .warden-rules.yml
# See docs/PARAMETERS.md for .warden-rules.yml format# Safe, educational review with explanations
warden \
--dry-run \
--reviewers generalist \
--verbose \
--save-report onboarding-feedback.md# Critical security fixes only
warden \
--reviewers security \
--review-focus security \
--test-strategy full \
--fix-strategy conservative \
--test-on-severity critical \
--comment-on-pr \
--notify-slack $SECURITY_CHANNEL# Focus on performance issues
warden \
--reviewers performance,architecture \
--review-focus performance \
--test-strategy smart \
--fix-strategy balanced \
--save-metrics .warden/perf-sprint-metrics.json# Address tech debt and maintainability
warden \
--reviewers maintainability,architecture \
--fix-strategy aggressive \
--test-strategy affected \
--max-fixes-per-tier 30# Comprehensive pre-release validation
warden \
--state all \
--reviewers security,performance,architecture,testing \
--test-strategy full \
--fix-strategy conservative \
--baseline-commit v1.2.0 \
--save-report release-candidate-$(date +%Y%m%d).md- QUICKSTART.md - 5-minute quick start
- PARAMETERS.md - Complete parameter reference
- WORKFLOW.md - Detailed workflow explanation
- SAFETY.md - Safety features and best practices