Patski123/coding-agent-examples
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

Add comprehensive README documentation for code and security review workflows

Browse Source 
- Add detailed README for code-review workflow with Anthropic claude-code-action reference
- Add detailed README for security-review workflow with claude-code-security-review reference
- Update root README with new workflow sections for code and security reviews
- Maintain consistent format across all workflow documentation
This commit is contained in:
Patrick Ellis
2025-09-14 00:08:20 -07:00
parent d176defdef
commit 45353885b7
3 changed files with 70 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
README.md
View File

@@ -5,6 +5,12 @@ Workflows are covered in detail with tutorials and demos on [Patrick Ellis' YouT
## Workflows ## Workflows
### [Code Review Workflow](./code-review/)
An automated code review system inspired by Anthropic's own Claude Code development process, where AI agents handle the "blocking and tackling" of code review. This workflow implements dual-loop architecture with slash commands and GitHub Actions to automatically review PRs for syntax, completeness, style guide adherence, and bug detection. Free your team to focus on strategic thinking and architectural alignment while AI handles routine checks. [Watch the tutorial](https://www.youtube.com/@PatrickOakleyEllis).
### [Security Review Workflow](./security-review/)
An automated security review system that proactively identifies vulnerabilities, exposed secrets, and potential attack vectors in your codebase. Based on Anthropic's security-focused approach and OWASP Top 10 standards, this workflow provides severity-classified findings with clear remediation guidance. Includes slash commands for on-demand scanning and GitHub Actions for automated PR security checks. [Watch the tutorial](https://www.youtube.com/@PatrickOakleyEllis).
### [Design Review Workflow](./design-review/) ### [Design Review Workflow](./design-review/)
An automated design review system that provides comprehensive feedback on front-end code changes. This workflow uses Microsoft's open source [Playwright MCP](https://github.com/microsoft/playwright-mcp) browser automation and specialized Claude Code agents to ensure UI/UX consistency, accessibility compliance, and adherence to world-class design standards. Perfect for maintaining design quality across teams and catching visual issues before they reach production. An automated design review system that provides comprehensive feedback on front-end code changes. This workflow uses Microsoft's open source [Playwright MCP](https://github.com/microsoft/playwright-mcp) browser automation and specialized Claude Code agents to ensure UI/UX consistency, accessibility compliance, and adherence to world-class design standards. Perfect for maintaining design quality across teams and catching visual issues before they reach production.

33
code-review/README.md Normal file
View File

@@ -0,0 +1,33 @@
# Code Review Workflow
This directory contains templates and examples for implementing an automated code review system that provides comprehensive feedback on code changes. This workflow, inspired by Anthropic's own Claude Code development process and their [claude-code-action](https://github.com/anthropics/claude-code-action) GitHub repository, enables teams to scale code review capacity while maintaining high quality standards through AI-assisted reviews.
## Concept
This workflow establishes a comprehensive methodology for automated code reviews in Claude Code, replacing manual line-by-line reviews with intelligent AI agents that handle pattern matching and consistency checks:
**Core Methodology:**
- **Automated Code Reviews**: Deploy AI reviewers that handle the "blocking and tackling" of code review - syntax, completeness, style guide adherence, and bug detection
- **Dual-Loop Architecture**: Leverage both inner loop (slash commands, subagents) for iterative development and outer loop (GitHub Actions) for automated PR validation
- **Standards-Based Evaluation**: Enforce consistent code quality through pattern matching, fast analysis, and adherence to your team's specific coding standards
- **Human-AI Collaboration**: Free human reviewers to focus on high-level strategic thinking, architectural alignment, and business logic while AI handles routine checks
**Implementation Features:**
- **Claude Code Subagents**: Deploy specialized code review agents that preserve context and provide detailed analysis without consuming main thread tokens
- **Slash Commands**: Enable instant code reviews with `/review` that automatically analyzes recent commits or specified PRs
- **GitHub Actions Integration**: Fully automated reviewers that run on every PR, providing consistent feedback before human review
- **Customizable Review Criteria**: Tailor review standards to your organization's specific needs, architectural patterns, and coding conventions
- **Learning Opportunities**: Teams learn from AI-generated reviews, improving their understanding of best practices and common pitfalls
This approach, battle-tested by Anthropic's own engineering team building Claude Code with Claude Code, enables teams to handle the increased volume of AI-generated code while maintaining rigorous quality standards.
## Resources
### Templates & Examples
- [Claude Code Review YAML](./claude-code-review.yml) - Standard GitHub Action configuration for automated code reviews
- [Custom Code Review YAML](./claude-code-review-custom.yml) - Extended configuration with custom review criteria
- [Pragmatic Code Review Slash Command](./pragmatic-code-review-slash-command.md) - Custom slash command for on-demand pragmatic code reviews
- [Pragmatic Code Review Subagent](./pragmatic-code-review-subagent.md) - Subagent configuration for comprehensive code analysis
### Video Tutorial
For a detailed walkthrough of this workflow, watch the comprehensive tutorial on YouTube: [Patrick Ellis' Channel - Coming Soon](https://www.youtube.com/@PatrickOakleyEllis)

31
security-review/README.md Normal file
View File

@@ -0,0 +1,31 @@
# Security Review Workflow
This directory contains templates and examples for implementing an automated security review system that provides comprehensive vulnerability scanning and security analysis on code changes. This workflow is inspired by and taken from Anthropic's [claude-code-security-review](https://github.com/anthropics/claude-code-security-review) GitHub repository, enabling teams to proactively identify and address security issues before they reach production.
## Concept
This workflow establishes a comprehensive methodology for automated security reviews in Claude Code, leveraging AI agents to detect vulnerabilities and enforce security best practices:
**Core Methodology:**
- **Automated Security Scanning**: Deploy AI-powered security reviewers that identify vulnerabilities, exposed secrets, and potential attack vectors
- **OWASP-Based Analysis**: Follow industry-standard security frameworks including OWASP Top 10 to ensure comprehensive coverage
- **Severity Classification**: Automatically categorize findings by severity level (Critical, High, Medium, Low) with clear remediation guidance
- **False Positive Management**: Intelligent filtering to reduce noise and focus on real security issues
**Implementation Features:**
- **Slash Commands**: Enable instant security reviews with `/security-review` that analyzes recent changes for security vulnerabilities
- **GitHub Actions Integration**: Automated security scanning on every PR, with inline comments highlighting specific security concerns
- **Secret Detection**: Identify exposed API keys, credentials, and sensitive information before they're committed
- **Dependency Analysis**: Review third-party dependencies for known vulnerabilities and security risks
- **Custom Security Policies**: Configure organization-specific security requirements and compliance standards
This approach ensures that security is built into the development process from the start, catching vulnerabilities early when they're easiest and least expensive to fix.
## Resources
### Templates & Examples
- [Security Review Slash Command](./security-review-slash-command.md) - Default security review command from Anthropic (source: [claude-code-security-review](https://github.com/anthropics/claude-code-security-review))
- [Security YAML](./security.yml) - GitHub Action configuration for automated security scanning
### Video Tutorial
For a detailed walkthrough of this workflow, watch the comprehensive tutorial on YouTube: [Patrick Ellis' Channel - Coming Soon](https://www.youtube.com/@PatrickOakleyEllis)