ce-session-inventory

Discover session metadata from Claude Code, Codex, and Cursor repositories.

2|Updated May 8, 2026
One-click install
npx skills add https://github.com/xotong/claude-marketplace --skill ce-session-inventory-xotong
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ce-session-inventory
Source: https://github.com/xotong/claude-marketplace/tree/main/plugins/compound-engineering/skills/ce-session-inventory
Command: npx skills add https://github.com/xotong/claude-marketplace --skill ce-session-inventory-xotong

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Discover cross-platform session files across Claude Code, Codex, and Cursor and emit session metadata to support agent research and auditing without direct user queries.

Core Features & Use Cases

  • Cross-platform session discovery from Claude Code, Codex, and Cursor with a single entry point.
  • Metadata extraction including platform, timestamps (last_ts), session id, branch, cwd, and file size, emitted as JSONL for downstream processing.
  • Flexible usage for auditing, debugging agent history, and researching project activity; supports optional filters like cwd and keyword matching to boost relevance.

Quick Start

Execute the discovery and metadata extraction workflows for your repository to begin generating session metadata.

Frequently Asked Questions about ce-session-inventory

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

FAQPage Schema
How do I discover session histories across Claude Code, Codex, and Cursor for a specific repository?

Cross-platform session discovery is performed by running script-driven workflows that scan Claude Code, Codex, and Cursor session files for a given repository. The tool emits structured JSONL metadata containing platform, timestamps, session id, branch, cwd, and file size for downstream auditing.

What metadata is extracted from agent session files for debugging behavior?

Agent session metadata extraction captures platform, last_ts timestamps, session id, branch, cwd, and file size. This structured JSONL output enables debugging agent behavior and auditing session histories across Claude Code, Codex, and Cursor platforms.

Can I filter session metadata by working directory or keyword when auditing project activity?

Optional cwd and keyword matching filters are supported through included scripts to boost relevance during session metadata discovery. Filtering helps narrow cross-platform project activity research to specific working directories or terms within Claude Code, Codex, and Cursor histories.

Does this session inventory tool work without dependencies or external libraries?

The session inventory tool operates with no external dependencies, using only included scripts for deterministic discovery. It requires no prerequisite packages to extract cross-platform session metadata from Claude Code, Codex, and Cursor into JSONL format.

What is the best way to audit cross-platform agent session histories for research?

Auditing cross-platform agent session histories is best handled by deterministic script-driven discovery that scans Claude Code, Codex, and Cursor simultaneously. The approach outputs structured JSONL session metadata, enabling efficient research and debugging without manual file searches.

Why use JSONL format for cross-platform session metadata extraction?

JSONL format supports structured, line-delimited session metadata emission for downstream processing and analysis. Each line contains platform, timestamps, session id, branch, cwd, and file size, making it ideal for auditing agent behavior across Claude Code, Codex, and Cursor.