dig

Mine Claude Code jsonl session files to reconstruct timelines and attribute work to repositories.

1|Updated May 24, 2026
One-click install
npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill dig-tinner-deinno
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: dig
Source: https://github.com/tinner-deinno/innova-skills-lib/tree/main/core/dig
Command: npx skills add https://github.com/tinner-deinno/innova-skills-lib --skill dig-tinner-deinno

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Digging through past Claude Code activity to reconstruct what you worked on, identify gaps, and attribute work to the correct repositories is usually slow and unreliable.

Core Features & Use Cases

  • Session goldmining: Mines Claude Code session history and outputs session data for timelines, gaps, and repo attribution.
  • Single-repo vs all-repos coverage: Supports scanning the current project repository or auto-detecting and scanning across all repos.
  • Timeline views with gap detection: Produces a chronological, day-by-day grouped timeline and highlights “sleeping/offline” and long gaps between sessions.
  • Deep historical scanning: A deep mode scans all jsonl session files to recover older, subagent, and worktree sessions that standard deduplication can miss.

Quick Start

Ask the assistant to show your most recent Claude Code sessions by saying: /dig.

Frequently Asked Questions about dig

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

FAQPage Schema
How do I reconstruct a timeline from my Claude Code session history?

To reconstruct a timeline from Claude Code session history, you can mine the jsonl session files to extract chronological activity, detect timezones for correct rendering, and output structured tables grouped by day. This recovers past work across repositories.

Can I track gaps and offline periods in my Claude Code sessions?

Yes, gap analysis in Claude Code sessions highlights sleeping, offline, and long periods between activity. By scanning session history chronologically, the tool identifies and visualizes these gaps within the generated day-by-day timeline.

How do I attribute past Claude Code work to specific repositories?

To attribute work to repositories, the session mining process scans Claude Code history to map session data to the corresponding repos. You can scan a single project repository or auto-detect and scan across all repos for full coverage.

Does Claude Code store session history in jsonl files for mining?

Yes, Claude Code stores session data in jsonl files. Mining these jsonl session files allows you to extract past activity, and a deep scanning mode can recover older, subagent, and worktree sessions that standard deduplication might miss.

What is the best way to review what I did previously across multiple repos in Claude Code?

The best way to review work across multiple repos is to auto-detect and scan all Claude Code repositories. This produces a comprehensive timeline with gap detection and repo attribution, ensuring full historical coverage of your sessions.

Why does my Claude Code session timeline show incorrect timestamps?

Incorrect timestamps in a Claude Code timeline usually result from missing timezone detection. Applying timezone detection during jsonl mining ensures correct time rendering for accurate chronological logging.