Overview

This skill triggers the smoke-test agent through the Agent tool to perform a quick pipeline verification. It returns the agent’s output verbatim (no commentary or formatting) so you can confirm basic health before deeper testing.

How This Skill Works

Initiate a call to the smoke-test agent by specifying subagent_type: claude-of-alexandria:smoke-test through the Agent tool. The agent executes its predefined checks and returns the raw output exactly as produced, with no extra headers or formatting.

When to Use It

• When you are asked to run a smoke test or pipeline verification.
• After a new build or pipeline change that requires quick sanity checks.
• Before committing to full integration tests to catch obvious failures.
• After provisioning or scaling infrastructure to validate basic readiness.
• When validating a plugin or integration update within Claude‑of‑alexandria.

Quick Start

1. Step 1: Prepare invocation payload with subagent_type: claude-of-alexandria:smoke-test.
2. Step 2: Run the Agent tool to invoke the smoke-test agent.
3. Step 3: Return the output verbatim with no commentary or formatting.

Best Practices

• Use the exact subagent_type claude-of-alexandria:smoke-test to trigger the test.
• Do not add commentary or formatting to the returned output; treat it as truth.
• Capture and store the raw output for debugging and auditing.
• Ensure environment permissions and network access are in place before invocation.
• If the output indicates issues, perform targeted deeper tests rather than re-running smoke-test blindly.

Example Use Cases

• CI pipeline smoke test after a new commit.
• Health check immediately after a fresh environment provision.
• Smoke-test as a quick gate before full E2E tests.
• Sanity check after updating a microservice.
• Verifying basic endpoints in a staging deployment.

Frequently Asked Questions