cli-debug-log-analysis

Community

Parse Ralph CLI logs to map subagent spans.

Authorskurekjakub
Version1.0.0
Installs0

System Documentation

What problem does it solve?

Parse and analyze Ralph CLI debug logs (cli-debug.log) to extract subagent spans, tool call sequences, token consumption, context compaction events, and error patterns. Use this skill when you need to manually parse a cli-debug.log file — for example when the subagent-mapper's pre-extracted data is insufficient or missing, when debugging a specific subagent's behavior in detail, or when investigating infrastructure issues visible only in raw logs. Trigger on phrases like 'parse the debug log', 'extract subagent spans', 'analyze tool calls from the log', 'what happened in the cli-debug log', or 'dig into the raw log'.

Core Features & Use Cases

  • Subagent span extraction: identify the start and end of each subagent invocation and pair them with the resolved model.
  • Tool call sequencing: map function/tool calls within each subagent span to reconstruct the execution trace.
  • Telemetry consolidation: capture token usage and context window metrics (e.g., context compaction events) for performance analysis.
  • Debugging aids: provide targeted extraction recipes to investigate infrastructure issues, model fallbacks, or tool failures.

Quick Start

Analyze a cli-debug.log to map subagent spans, tool calls, and token usage for a given run.

Dependency Matrix

Required Modules

None required

Components

Standard package

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: cli-debug-log-analysis
Download link: https://github.com/skurekjakub/docwriter-agent/archive/main.zip#cli-debug-log-analysis

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.