cardscan-ai/Claude Code Watchdog

Your AI watchdog that watches for test failures and heals them automatically

View on GitHub

Trust Signals

Scorecard Score
not yet scored
Maintenance Recency
Maintainedlast commit Dec 20, 2025
License
MIT

Pinned Snippet

workflow.ymlSHA-pinned
uses: cardscan-ai/claude-code-watchdog@00b5ba76f5f1e40a7d21f71f072149299bcb4a63 # v0.3.1

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

namedescriptionrequireddefault
anthropic_api_keyAnthropic API key for Claudeyes
severity_thresholdMinimum severity to process (ignore|low|medium|high|critical)nomedium
create_issuesCreate GitHub issues for failuresnotrue
create_fixesAttempt to implement fixes automaticallynotrue
rerun_testsAttempt to re-run failed tests to confirm fixesnofalse
debug_modeUpload watchdog artifacts for debuggingnofalse
safe_modeSkip potentially risky external content (GitHub issues, PRs, commit messages)nofalse
test_results_pathPath or glob pattern to test result files (e.g., "test-results/**/*.xml", "cypress/reports/*.json")yes
namedescription
severityFailure severity (ignore|low|medium|high|critical)
action_takenWhat action was taken (issue_created|issue_updated|pr_created|pr_updated|tests_fixed|none)
issue_numberGitHub issue number if created or updated
pr_numberPR number if fixes were created
tests_passingtrue if re-run tests passed after fixes