amondnet/Claude Code Review

AI-powered comprehensive code review using Claude with customizable review types and automated feedback

Trust Signals

workflow.ymlSHA-pinned

uses: amondnet/claude-code-review@a31378a4c98374cd8f67a4b1603683f23cbfff29 # no releases — HEAD as of 2026-07-13

tags can be moved; commit SHAs can't. why a SHA?

name	description	required	default
anthropic_api_key	Anthropic API key (required for direct API, not needed for Bedrock/Vertex)	no	—
claude_code_oauth_token	Claude Code OAuth token (alternative to anthropic_api_key)	no	—
github_token	GitHub token with repo and pull request permissions (optional if using GitHub App)	no	${{ github.token }}
use_bedrock	Use Amazon Bedrock with OIDC authentication instead of direct Anthropic API	no	false
use_vertex	Use Google Vertex AI with OIDC authentication instead of direct Anthropic API	no	false
review_type	Type of review to perform (comprehensive, security, performance, custom)	no	comprehensive
custom_prompt	Custom review prompt for specialized review needs	no	—
files_to_review	Specific files or patterns to review (default: all changed files)	no	—
severity_labels	Automatically add severity labels to PR based on issues found	no	true
track_progress	Force tag mode with tracking comments	no	false
exclude_paths	Comma-separated list of paths to exclude from review	no	node_modules,dist,build,.git
max_review_comments	Maximum number of inline comments to generate	no	50

name	description
review_summary	Summary of the code review results
issues_found	Total number of issues found
critical_issues	Number of critical severity issues
high_issues	Number of high severity issues
medium_issues	Number of medium severity issues
low_issues	Number of low severity issues