samushi/AI Code Review Agent

AI-powered, auto-stack code reviews for PRs (React/Next, Vue/Nuxt, Laravel/WordPress, Django/Flask …)

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
github-tokenGitHub token with repo scope (PR comments)yes${{ github.token }}
providerLLM provider: openai | anthropic | gemini | ollamanoopenai
openai-api-keyOpenAI API key (if provider = openai)no
anthropic-api-keyAnthropic Claude API key (if provider = anthropic)no
gemini-api-keyGoogle Generative AI key (if provider = gemini)no
ollama-base-urlBase URL of your Ollama server (if provider = ollama)nohttp://localhost:11434
ai-modelModel name for the chosen providernogpt-4o-mini
stackForce stack (react, vue, laravel, django, …) or leave blank for auto-detectno
file-patternsComma-separated glob patterns to includeno**/*.{js,jsx,ts,tsx,vue,php,blade.php,py,json,yml}
exclude-patternsComma-separated glob patterns to ignoreno**/{node_modules,dist}/**
min-score-thresholdMinimum score before failing CI (with fail-on-low-score = true)no7
post-commentWhether to post the AI report as a PR commentnotrue
fail-on-low-scoreFail the job when review-score < min-score-thresholdnofalse
namedescription
review-scoreNumeric score (1-10)
recommendationPOSITIVE | NEEDS_CHANGES | NEGATIVE
issues-foundNumber of findings reported by the AI
review-summaryOne-paragraph summary of the review