skills.rest
junghan0611
junghan0611COMMUNITY
Explore all skills by this author →
Website𝕏@junghanacsGitHub

session-recap

Extract user and assistant text from pi and Claude Code session JSONL.

6|Updated Nov 14, 2025
Search Tags:#context#session#summary#jsonl#recap#multiharness
One-click install
npx skills add https://github.com/junghan0611/agent-config --skill session-recap-junghan0611
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: session-recap
Source: https://github.com/junghan0611/agent-config/tree/main/skills/session-recap
Command: npx skills add https://github.com/junghan0611/agent-config --skill session-recap-junghan0611

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

The session-recap skill extracts only user and assistant text from a session JSONL to quickly restore context across multi-harness environments. It trims JSON noise and reduces token usage by focusing on the essential dialogue.

Core Features & Use Cases

  • Extracts only user/assistant messages from session logs (JSONL) to produce concise context.
  • Supports multi-harness environments (pi and Claude Code) with optional project filtering and session limits.
  • Enables fast context restoration across sessions for continuity in AI agent workflows.

Quick Start

Run session-recap to extract a concise summary from the most recent session(s) for a given project.

Frequently Asked Questions about session-recap

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

FAQPage Schema
How do I extract user and assistant messages from Claude Code session JSONL files?▼

To extract user and assistant messages from Claude Code session JSONL files, this skill parses the logs, trims JSON noise, and outputs structured dialogue text to restore context efficiently.

Can I summarize session logs across multiple AI harness environments like pi and Claude Code?▼

Yes, you can summarize session logs across multiple harness environments. It parses both pi and Claude Code JSONL formats to extract essential dialogue while filtering out background noise.

What is the best way to restore context from previous AI chat sessions without exceeding token limits?▼

The best way to restore context without exceeding token limits is to summarize session JSONL files. This skill extracts only user and assistant text, reducing token usage by focusing on essential dialogue.

How do I filter session summaries by project or limit output to the last N sessions?▼

You can filter session summaries by project or limit output to the last N sessions using the skill's built-in filtering options. It processes the specified JSONL logs to produce concise context restoration.

Does session recap output include commit history and cost data from the JSONL logs?▼

Yes, session recap output can include commit history and cost data. The skill parses JSONL logs to output structured data containing extracted messages alongside optional commits and cost metrics.

Why does my session context restoration include so much irrelevant JSON noise?▼

Session context restoration includes JSON noise because raw logs contain metadata. This skill trims that noise by extracting only user and assistant text to provide concise, structured dialogue output.

Related Skills

Explore More Content & Communication Skills→
openclawopenclaw
1.0k

session-viewer

Convert JSONL session transcripts from Codex, Claude Code, OpenClaw, or Pi into interactive HTML files.

Official
Intermediate
shishishishi
3

scrum-team-scrum-master

Coordinate Scrum ceremonies and update scrum.ts as the single source of truth.

Community
Basic
aresbitaresbit
46

session-logs

Search and analyze historical session JSONL files using jq and rg.

Community
Intermediate
codingSamsscodingSamss
11

orbit-session-diary

Convert Codex and Claude session JSONL logs into Markdown daily diary entries.

Community
Intermediate