Agent skill · operations · jeremylongshore
running-smoke-tests
Execute fast smoke tests validating critical functionality after deployment.
Why this skill is useful
Adds a comprehensive smoke test script and configuration patterns that validate critical application functionality post-deployment, which the AI wouldn't reliably generate on its own.
What it needs
Requires curl, node-fetch, playwright, wget installed locally. About 3k tokens when loaded. Last updated 2026-08-07. 2,604 stars on the source repository.
What this skill does
Smoke Test Runner Overview Execute fast, high-confidence smoke tests that validate critical application functionality after deployment or build. Smoke tests verify that the application starts, core user flows work, and key integrations respond -- without running the full test suite. Prerequisites Application deployed and accessible at a known URL or running locally HTTP client available (curl, wget, node-fetch, or Playwright) List of critical endpoints and user flows to validate Expected response codes and content patterns for each check CI/CD pipeline hook for post-deployment validation Instructions 1. Identify the critical paths that constitute a "working" application: Health check endpoint returns 200 with expected body. Homepage loads and contains key UI elements. Authentication flow succeeds with test credentials. Primary API endpoint returns valid data. Database connection is active and responding. 2. Create a smoke test configuration listing each check: URL or command to execute. Expected HTTP status code (200, 301, etc.). Response body pattern to match (substring or regex). Maximum acceptable response time (e.g., 3 seconds). 3. Write the smoke test suite as a lightweight script or test file: Use curl for HTTP checks or Playwright for browser-based checks. Run checks sequentially for simplicity (parallel for speed if independent). Fail fast on the first critical failure. Log each check result with pass/fail, response time, and status code. 4. Implement timeout guards: Set a global timeout of 60 seconds for the entire smoke suite. Set per-check timeouts of 5-10 seconds. Treat timeouts as failures, not retries. 5. Add deployment-gate integration: On success: proceed with deployment promotion or traffic shifting. On failure: trigger rollback and send alert notification. Report results to CI/CD dashboard and Slack/Teams webhook. 6. Store smoke test results as CI artifacts for audit trail. 7. …
How to use it
Reference it in AdaL, Claude Code, Cursor or any coding agent — nothing to install:
@skills jeremylongshore/running-smoke-tests--c24233