analyze-subagent-logs

Analyze subagent response logs for format violations, permission issues, and scope drift.

2|Updated Jan 9, 2024
One-click install
npx skills add https://github.com/tetran/Hotwire-ToDo --skill analyze-subagent-logs
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: analyze-subagent-logs
Source: https://github.com/tetran/Hotwire-ToDo/tree/main/.claude/skills/analyze-subagent-logs
Command: npx skills add https://github.com/tetran/Hotwire-ToDo --skill analyze-subagent-logs

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Analyze subagent response logs (.claude/logs/subagent_responses.jsonl) to find format violations, permission issues, scope drift, and other patterns worth addressing. Manual invocation only.

Core Features & Use Cases

  • Identify format violations and permission issues in subagent responses.
  • Surface recurring patterns to inform improvements to agent definitions, hooks, and delegation payloads.
  • Provide a concise manual-audit workflow for quality assurance of rails-developer and react-developer interactions.

Quick Start

Run the skill on the target log file at .claude/logs/subagent_responses.jsonl to generate a findings report.

Frequently Asked Questions about analyze-subagent-logs

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

FAQPage Schema
How do I analyze subagent response logs for format violations and permission issues?

You can identify recurring patterns by running a manual audit on the subagent_responses.jsonl log file, which surfaces scope drift and permission issues to inform improvements to agent definitions and hooks.

What is subagent scope drift and how do I detect it in JSONL logs?

Subagent scope drift occurs when an agent exceeds its intended task boundaries. You detect it by running a log analysis on subagent_responses.jsonl, which outputs per-entry diagnostics highlighting deviations.

How do I audit rails-developer and react-developer subagent interactions?

Audit rails-developer and react-developer interactions by running a manual quality assurance workflow on the subagent_responses.jsonl log file, which generates a structured summary and per-entry diagnostics for review.

Can I use manual log analysis to improve agent definitions and delegation payloads?

Yes, manual log analysis identifies recurring issues in subagent responses and outputs a concise recommendations list that directly informs improvements to agent definitions, hooks, and delegation payloads.

What do I need to generate a subagent log findings report?

You need access to the .claude/logs/subagent_responses.jsonl file. Running the skill on this target log file generates a findings report containing a structured summary, per-entry diagnostics, and recommendations.