adriangohjw/Saltman

Analyzes PR content and posts results as a comment

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Activelast commit Feb 8, 2026
License
None

Pinned Snippet

workflow.ymlSHA-pinned
uses: adriangohjw/saltman@5b6587c099a6b3570a94877bde224e1b5abd3216 # v1.1.0

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

namedescriptionrequireddefault
github-tokenGitHub token for API accessyes
providerLLM provider to use for code review. Must be either "openai", "anthropic", or "openai-compatible". OpenAI is recommended.yes
api-keyAPI key for the specified LLM provider.yes
base-urlBase URL for OpenAI-compatible API. Required when provider is "openai-compatible".no
modelModel name to use. For OpenAI: gpt-5.1-codex-mini (default), gpt-5.1-codex-max, or gpt-5.2-codex. For Anthropic: claude-opus-4-5 (default), claude-sonnet-4-5 or claude-haiku-4-5. Required when provider is "openai-compatible".no
post-comment-when-no-issuesWhether to post the analysis as a comment on the PR when no issues are detected (true/false). Defaults to false if not specified.no
ignore-patternsNewline-separated list of glob patterns to ignore files from PR analysis (e.g., "**/*.test.ts\n**/node_modules/**\n*.md"). Similar to .eslintignore or .gitignore patterns.no
target-branchBranch name to monitor for direct pushes. When set and action is triggered on push event, creates a GitHub issue instead of PR comments. Only used when triggered on push events.no
ping-usersSpace or newline-separated list of GitHub usernames or teams to ping (e.g., "@team/security @user1 @user2"). All items must start with "@".no
severity-filterNewline-separated list of severity levels to include in the review (e.g., "critical\nhigh\nmedium"). Valid values: critical, high, medium, low, info. If unspecified, all severities are shown.no

no outputs