ce-session-extract

Extract conversation skeletons and error signals from Claude Code, Codex, or Cursor session files.

Updated May 4, 2026
One-click install
npx skills add https://github.com/IAmKongHai/compound-engineering-plugin-zh --skill ce-session-extract-iamkonghai
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ce-session-extract
Source: https://github.com/IAmKongHai/compound-engineering-plugin-zh/tree/main/3.4.2/skills/ce-session-extract
Command: npx skills add https://github.com/IAmKongHai/compound-engineering-plugin-zh --skill ce-session-extract-iamkonghai

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This Skill automates the analysis of session files to detect errors and extract conversation skeletons, reducing manual analysis time and improving efficiency.

Core Features & Use Cases

  • Error Detection: Automatically identifies and logs errors from Claude Code, Codex, or Cursor session files.
  • Conversation Skeleton Extraction: Generates a narrative-readable digest of user messages, assistant text, and tool call summaries.
  • Use Case: Ideal for session research agents to quickly review and identify issues or key insights from large sets of session files.

Quick Start

Run the ce-session-extract skill on the session file located at '/path/to/session.jsonl' and specify 'errors' mode to get error signals or 'skeleton' mode to get a conversation summary.

Frequently Asked Questions about ce-session-extract

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

FAQPage Schema
How do I extract conversation skeletons from JSONL session files?

To extract conversation skeletons from JSONL session files, run a Python script that parses Claude Code, Codex, or Cursor logs to generate a readable digest of user messages, assistant text, and tool call summaries.

What is the best way to detect errors in Claude Code or Cursor session logs?

The best way to detect errors in Claude Code or Cursor session logs is to run an automated Python script in errors mode to scan JSONL files, identify issues, and automatically log the error signals.

Can I use Python scripting to automate session analysis for large sets of files?

Yes, you can use Python scripting to automate session analysis for large sets of JSONL files, allowing session research agents to quickly review and identify issues or key insights without manual inspection.

Do I need to specify a mode when processing session files for error detection?

Yes, you need to specify either 'errors' mode to get error signals or 'skeleton' mode to get a conversation summary when processing your session files for automated review.

What does a conversation skeleton contain when extracted from session files?

A conversation skeleton extracted from session files contains a narrative-readable digest that summarizes user messages, assistant text, and tool call summaries into a concise overview.