project-health - SKILL.md Agent Skill

name: project-health description: Monitor Semaphore project health — pass rates, recent failures, deployment status, trends. user-invocable: false

Project Health Monitoring

Quick health check

sem-ai health --project my-app

Returns: verdict (healthy/degraded/unhealthy), pass rate, failed/passed/other counts, deploy target count.

CI status for a branch

sem-ai status --project my-app --branch main
sem-ai status --project my-app --pr 422
sem-ai status   # auto-detects project + branch from git

status pins the current HEAD commit (look for "matched_by":"commit_sha"). Add --exit-code for a poll-friendly loop (0=pass / 8=pending / 1=fail / 2=ambiguous / 3=none): until sem-ai status --exit-code; do sleep 20; done watches to green. Prefer this over gh pr checks for the is-it-green check — same Semaphore status, and the failure drill (sem-ai diagnose <workflow-id>) is one tool away.

Recent workflows

sem-ai workflow list --project my-app
sem-ai workflow list --project my-app --branch main

Pipeline details

sem-ai pipeline list --project my-app    # all recent pipelines
sem-ai pipeline show <id>                # blocks + jobs tree

Test trends

sem-ai test flaky --project my-app --count 10   # flaky tests across last 10 workflows
sem-ai test summary --pipeline <id>              # test results for specific run

Deployment status

sem-ai deploy targets --project my-app
sem-ai deploy history <target-id>

Interpreting health verdict

healthy: pass rate >= 80%
degraded: pass rate 50-80%
unhealthy: pass rate < 50%