merkuryo/AI Code Reviewer

Automatically reviews PRs using Claude or a local Ollama LLM

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
llm_providerLLM provider to use: claude or ollamanoollama
anthropic_api_keyAnthropic API key (required if llm_provider=claude)no""
ollama_urlOllama API URLnohttp://localhost:11434
modelModel name (e.g. llama3.2, claude-3-5-haiku-20241022)nollama3.2
max_filesMaximum number of files to review (controls costs)no10
focus_areasComma-separated focus areas: bugs,security,performance,stylenobugs,security,performance
post_commentsWhether to post review comments on the PRnotrue
min_severityMinimum severity to comment on: low, medium, high, criticalnomedium
namedescription
review_summarySummary of the review findings
issues_foundNumber of issues found