Prompt Log for Openclaw

A utility to transform raw JSONL session logs from AI coding agents into readable Markdown transcripts.

thesash
v0.1.0
Jan 13, 2026
3
4.4k
0

Install & Download

1. ClawHub CLI

The fastest way to install a skill directly from the registry.

npx clawhub@latest install prompt-log

2. Manual Installation

Copy the skill folder to one of these locations

Global
~/.openclaw/skills/
Workspace
<project>/skills/

Priority: Workspace > Local > Bundled

3. Prompt Installation

Copy this prompt to OpenClaw to install it automatically.

Help me install prompt-log using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).

Prefer to download?

Get the raw skill files in a ZIP archive.

What is Prompt Log?

Prompt Log is a specialized utility designed for developers who need to bridge the gap between raw execution data and readable documentation. It extracts conversation transcripts from .jsonl session files generated by popular AI coding assistants such as Clawdbot, Claude Code, and Codex. By utilizing this utility within Openclaw Skills, developers can ensure their AI interactions are preserved in a format that is easy to read, share, and archive.

This skill is particularly useful for maintaining a clear audit trail of AI-assisted development. It processes complex log structures into clean Markdown files, making it an essential part of the Openclaw Skills ecosystem for anyone looking to document their prompt engineering workflows or session histories.

Prompt Log Use Cases

  • Archiving AI coding sessions for project documentation.
  • Exporting prompt history from Clawdbot or Codex into readable formats for team reviews.
  • Filtering long session logs to find specific interactions based on ISO timestamps.
  • Generating Markdown transcripts for technical blog posts or internal wikis using Openclaw Skills.

How Prompt Log Works

  1. Point the script to a .jsonl session log file generated by an AI coding agent.
  2. The utility uses jq to parse the structured JSON data and isolate message content.
  3. Optional filters like --after and --before are applied to refine the output based on specific times.
  4. The processed data is formatted into a clean Markdown structure for readability.
  5. The final transcript is saved to the designated .prompt-log directory or a custom path defined by the user of Openclaw Skills.

Prompt Log Setup

To get started with this skill, ensure you have the necessary dependencies and run the extraction script:

# Ensure jq is installed in your PATH
# On macOS, use Homebrew:
brew install jq

# Run the extraction script on a session file
scripts/extract.sh <session-file>

Prompt Log Data Schema & Taxonomy

The Prompt Log skill organizes data based on the source session and user-defined parameters:

Attribute Specification
Input Format .jsonl (Clawdbot, Claude Code, or Codex logs)
Output Format Markdown (.md)
Default Path .prompt-log/YYYY-MM-DD-HHMMSS.md
Filtering ISO 8601 timestamps for --after and --before
Logic Uses jq for parsing and date/gdate for time handling

Prompt Log Advanced Features

  • Temporal filtering support for granular session extraction.
  • Automatic directory management for storing transcripts in project-local folders.
  • Cross-platform date handling with gdate support for macOS users.
  • Flexible output redirection for seamless integration into other Openclaw Skills workflows.

SKILL.md


Loading

Related Openclaw Skills

METADATA

Github Stars: 0
forks: 0

Featured*