investigate-cron

Locate matching session transcripts and reconstruct Claude Code cron run behavior.

9|2|Updated Jun 6, 2026
One-click install
npx skills add https://github.com/app-vitals/shipwright --skill investigate-cron
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate-cron
Source: https://github.com/app-vitals/shipwright/tree/main/plugins/shipwright/skills/investigate-cron
Command: npx skills add https://github.com/app-vitals/shipwright --skill investigate-cron

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill helps you understand why a Claude Code cron run behaved the way it did by tracing the matching session transcript, instead of guessing from incomplete logs.

Core Features & Use Cases

  • Transcript matching: Finds the most likely session by cron name and approximate execution time.
  • Behavior reconstruction: Extracts the assistant’s outputs and shell commands to show what the run actually did.
  • Fallback handling: Explains when no matching session exists and points you toward the next check.
  • Use Case: Use it when a deploy, review, or patch cron fired unexpectedly and you need a clear plain-language explanation of the outcome.

Quick Start

Ask the skill to investigate a specific cron by name and approximate time so it can locate the transcript and explain what happened.

Frequently Asked Questions about investigate-cron

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

FAQPage Schema
How do I investigate a Claude Code cron run from its session transcript?

Reconstructing a Claude Code cron run involves locating the matching session JSONL transcript by cron name and approximate firing time to extract assistant outputs and shell commands, explaining what the run actually did.

Why did my Shipwright deploy or patch cron fire unexpectedly?

An unexpected Shipwright cron run is diagnosed by tracing the matching session transcript to extract outputs and shell commands, reconstructing exactly what the deploy, review-patch, dev-task, patch, or review scenario executed.

What is the best way to troubleshoot Claude Code cron behavior using JSONL logs?

Troubleshooting Claude Code cron behavior via JSONL logs requires matching the session transcript to the cron name and Pacific-time firing window, then extracting the assistant's actions to reconstruct the run instead of guessing from incomplete logs.

Can I trace a Claude Code session transcript if I only have an approximate cron firing time?

Yes, tracing a Claude Code session transcript with an approximate firing time is possible by converting it to Pacific time and matching it against the cron name to locate the correct JSONL workspace path encoding.

What should I do when no matching session transcript is found for a cron investigation?

When no matching session transcript is found for a cron investigation, the process provides graceful fallback guidance that explains the missing session and points you toward the next diagnostic check for the unexpected run.