p2achai/Codex Reviewer

AI-powered PR review that provides summary, code improvement suggestions, and bug detection as a comment

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Stale
License
None
namedescriptionrequireddefault
github_tokenGitHub token used for posting review comments on PRsyes
providerAI provider to use: openai or claudenoclaude
openai_api_keyOpenAI API Key (required when provider=openai)no""
anthropic_api_keyAnthropic API Key (required when provider=claude)no""
labelLabel name that triggers the review action (e.g. codex-review)yescodex-review
spec_labelLabel for spec+tests reviewnocodex-review
perfsec_labelLabel for performance/security reviewnocodex-review-perf
bug_labelLabel for correctness/bug reviewnocodex-review-bug
high_labelLabel for high-quality review (uses premium models)nocodex-review-high
modelModel to use for code review (e.g. gpt-5-mini, o4-mini for OpenAI; claude-sonnet-4-6, claude-opus-4-6 for Claude)noclaude-opus-4-6
effortClaude Code effort level for code review (low, medium, high, xhigh, max)nohigh
codex_versionPinned @openai/codex version to installno0.115.0-alpha.27
claude_code_versionPinned @anthropic-ai/claude-code version to installno2.1.150
languageLanguage for the review output (e.g. english, korean, japanese, etc.)noenglish
custom_promptOptional custom prompt to override the default review promptno""
enable_multi_agentDeprecated. Multi-agent review is no longer used.nofalse
agents_pathDeprecated. Multi-agent agent definition path is no longer used.noagents.json
clickup_api_tokenClickUp API token for fetching spec/task contentno""
clickup_urlClickUp task URL for spec compliance agentno""
clickup_team_idClickUp team/workspace ID (required when using custom task IDs)no""
clickup_custom_task_idsWhether ClickUp uses custom task IDs (true/false)nofalse
spec_sourceSpec URL source: input, comment, or autonoauto
spec_comment_markerMarker used to find spec URL in PR commentsnoSPEC:

no outputs