extract-transcripts

Extract Claude Code and Codex session JSONL transcripts into Markdown.

53|7|Updated Jun 28, 2025
One-click install
npx skills add https://github.com/0xBigBoss/claude-code --skill extract-transcripts
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: extract-transcripts
Source: https://github.com/0xBigBoss/claude-code/tree/main/.claude/skills/extract-transcripts
Command: npx skills add https://github.com/0xBigBoss/claude-code --skill extract-transcripts

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

Extracts readable transcripts from Claude Code and Codex session JSONL files, enabling quick review, auditing, and knowledge reuse.

Core Features & Use Cases

  • Transcript extraction: Convert raw session JSONL into clean, readable Markdown per session.
  • Cross-session search: Build searchable transcripts to compare prompts, responses, and tool usage across sessions.
  • Auditing & documentation: Preserve model, working directory, and git context for traceability.
  • Use Case: Imagine you have 100 Claude Code sessions; extract them to a consistent Markdown format for faster review and compliance.

Quick Start

Run the extract-transcripts tool on a single JSONL session:

  • Claude Code: uv run ~/.claude/skills/extract-transcripts/extract_transcript.py <session.jsonl>
  • Codex CLI: uv run ~/.claude/skills/extract-transcripts/extract_codex_transcript.py <session.jsonl>

Frequently Asked Questions about extract-transcripts

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

FAQPage Schema
How do I convert Claude Code session JSONL files into readable Markdown?

To convert Claude Code session JSONL files into readable Markdown, you can run the extract_transcript.py Python script, which parses raw session inputs and generates clean Markdown output per session.

Can I extract Codex CLI transcripts into a searchable format?

Yes, you can extract Codex CLI transcripts into a searchable format by running the extract_codex_transcript.py script on your session JSONL files to generate markdown-friendly output.

Does transcript extraction preserve tool usage and thinking blocks?

Yes, transcript extraction optionally preserves tool usage and thinking blocks alongside message content and per-session metadata, ensuring full auditing and knowledge reuse capabilities.

What is the best way to audit multiple Claude Code sessions for compliance?

The best way to audit multiple Claude Code sessions is to extract them into a consistent Markdown format, preserving model, working directory, and git context for cross-session review and traceability.

Do I need any specific dependencies to run the transcript extraction scripts?

You do not need any external dependencies to run the transcript extraction scripts, as the implementation relies solely on Python scripts executed via the uv run command.