monitoring

Check Hawk job status and retrieve logs for diagnostics.

1|Updated May 12, 2018
One-click install
npx skills add https://github.com/tbroadley/dotfiles --skill monitoring-tbroadley
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: monitoring
Source: https://github.com/tbroadley/dotfiles/tree/main/claude/skills/hawk-monitoring
Command: npx skills add https://github.com/tbroadley/dotfiles --skill monitoring-tbroadley

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps operators quickly verify Hawk job status, access relevant logs, and identify failures without manual digging.

Core Features & Use Cases

  • Live status checks for Hawk evaluating runs, including the latest run.
  • Log retrieval and diagnostics for a failing or completed job.
  • Automated monitoring reports that summarize status, logs, and key metrics.

Quick Start

hawk status hawk logs hawk status <JOB_ID> > report.status

Frequently Asked Questions about monitoring

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I check Hawk job status for active and completed evaluation runs?

To check Hawk job status, run the command `hawk status` to retrieve live updates on active and completed evaluation runs. Omitting the JOB_ID automatically queries the last known eval_set_id or scan_run_id for quick visibility.

How do I retrieve Hawk logs to diagnose a failing evaluation job?

Retrieve Hawk logs by running the `hawk logs` command to diagnose failing jobs. Specify a JOB_ID to target a specific evaluation run, or access the latest run automatically if the ID is omitted.

Do I need the hawk CLI installed to monitor evaluation runs and generate reports?

Yes, you need the hawk CLI installed and available in your execution environment to monitor Hawk jobs. The Skill relies on this CLI to identify job status, pull relevant logs, and generate automated monitoring reports.

What's the best way to generate a monitoring report for a completed Hawk run?

The best way to generate a monitoring report is redirecting the `hawk status <JOB_ID>` command output to a file. This summarizes the run status, relevant logs, and key metrics for your evaluation diagnostics.

Why does my Hawk job monitoring command fail to find the correct evaluation run?

Hawk job monitoring commands fail if the hawk CLI is missing from the execution environment or the specified JOB_ID is invalid. When omitted, the command defaults to the last known eval_set_id or scan_run_id.