From github
Monitors GitHub Actions workflow runs, identifies failing jobs, extracts error logs, and summarizes diagnostics. Useful for checking CI status after pushes or investigating failures.
npx claudepluginhub bendrucker/claude --plugin githubThis skill is limited to using the following tools:
Monitor GitHub Actions and extract failure diagnostics. Extract only: find the failures and present them concisely. Do not analyze root causes or suggest fixes.
Creates isolated Git worktrees for feature branches with prioritized directory selection, gitignore safety checks, auto project setup for Node/Python/Rust/Go, and baseline verification.
Executes implementation plans in current session by dispatching fresh subagents per independent task, with two-stage reviews: spec compliance then code quality.
Dispatches parallel agents to independently tackle 2+ tasks like separate test failures or subsystems without shared state or dependencies.
Monitor GitHub Actions and extract failure diagnostics. Extract only: find the failures and present them concisely. Do not analyze root causes or suggest fixes.
$ARGUMENTS
If no run ID or PR is specified, use the current branch.
!git branch --show-current
Fetch recent runs for the current branch:
gh run list --branch <current-branch> --limit 5 --json databaseId,status,conclusion,displayTitle,workflowName
Query the PR to get the source branch SHA: gh pr view --json headRefOid. Report this as "Source SHA" in the output. Compare against the run's headSha to detect merge queue or other synthetic commit scenarios.
From the recent runs, identify the most relevant run (latest, or matching the target). If the run is still in progress, use gh run watch <id> to wait for completion.
gh run view <run-id> --json jobs --jq '[.jobs[] | select(.conclusion == "failure") | {name, databaseId, conclusion}]'
For each failing job, fetch its log:
gh run view --log --job <job-id>
Parse the log output for failure-relevant sections. See references/log-parsing.md for CI-specific log structure.
If the full job log is still too large, try --log-failed on the specific job:
gh run view --log-failed --job <job-id>
When multiple jobs fail, fetch logs for each job in separate parallel Bash calls.
For each failed job, report:
Keep the total output concise. The parent conversation will use this to investigate.