🗂️
AI EngineeringPython

Harness Step 3 Session Management

by simbajigege

Harness Step 3 Session Management is an AI Engineering skill for Claude Code, published by simbajigege in book2skills.

161 stars30 forkson simbajigege/book2skillsAdded 2026/09/08+4% in starsRepository updated 2026/08/26
agent-skillsagentskillsanthropicanthropic-claudebook2skillsgrowth-investinginvestinginvesting-skillsskillsstock-analysis
Install in seconds
Install Harness Step 3 Session Management
Copy Harness Step 3 Session Management into your Claude Code skills folder. Run the command in your terminal, or review the source on GitHub before installing.
terminal
npx degit https://github.com/simbajigege/book2skills/tree/main/skills/harness-step3-session-management ~/.claude/skills/harness-step3-session-management

Requires Node.js. Downloads this skill only — not the rest of the repository — into your Claude Code skills folder.

Without Node.js

git clone https://github.com/simbajigege/book2skills.git

Clones the whole repository, then copy the skill’s own directory into your skills folder yourself.

In this catalog

Source file
skills/harness-step3-session-management/SKILL.md in simbajigege/book2skills
Installs to
~/.claude/skills/harness-step3-session-management
Collection
One of 25 skills cataloged from this repository
Category
AI Engineering3670 skills

What Harness Step 3 Session Management does

Harness Step 3 Session Management creates init.sh, tasks.json, and progress.md to carry work state across sessions. Use it when an agent needs to resume tasks, track progress, and verify the environment quickly.

Harness Step 3 Session Management is cataloged under AI Engineering on DirSkills. Harness Step 3 Session Management comes from a repository tagged agent-skills, agentskills, anthropic, anthropic-claude and book2skills.

Documentation

README

Harness Step 3: 建立跨 Session 状态管理

目标

创建三个文件,让 agent 在任何新 session 开始时能在 30 秒内恢复工作状态:

  • init.sh:环境初始化脚本,验证项目可以正常启动
  • tasks.json:当前任务清单,agent 的工作指令来源
  • progress.md:人类可读的进度摘要,记录每次 session 的关键信息

核心原则:状态靠文件传递,不靠 agent 的记忆。git log 是主记录,这三个文件是辅助。


执行步骤

Step 1:扫描项目启动方式

在写 init.sh 之前,先确认项目如何启动和测试:

# 读 package.json 的 scripts(Node.js 项目)
cat package.json 2>/dev/null | grep -A 20 '"scripts"'

# 或读 Makefile(多语言项目)
cat Makefile 2>/dev/null | head -40

# 或读 pyproject.toml(Python 项目)
cat pyproject.toml 2>/dev/null | grep -A 20 '\[tool.poetry.scripts\]'

# 确认现有 AGENTS.md 里的启动命令
grep -A 5 '启动命令\|start\|dev\|run' AGENTS.md 2>/dev/null

收集:

  • 开发服务器启动命令
  • 测试命令
  • 类型检查/lint 命令(如果有)
  • 有没有需要先跑的初始化步骤(如数据库迁移)

Step 2:创建 init.sh

This is the opening of the README. Read the full README on GitHub.

Frequently asked about Harness Step 3 Session Management

  • What else does simbajigege publish alongside Harness Step 3 Session Management?

    Harness Step 3 Session Management is one of 25 skills that DirSkills catalogs from simbajigege/book2skills, the repository it ships in. Its siblings there include Agent Memory Implementation, Agent Tool Builder and Business Adventures Analysis. Each one is a separate skill with its own page in this directory, installs the same way Harness Step 3 Session Management does, and is maintained by simbajigege in that same repository. The rest of the collection is listed on the simbajigege/book2skills page.

  • How does Harness Step 3 Session Management compare to other AI Engineering skills?

    Harness Step 3 Session Management ranks #3434 by stars among the 3670 AI Engineering skills in this catalog. The most-starred ones next to it are Architecture Decision Records, AI-First Engineering and Agentic OS. DirSkills ranks by the star count of the repository each skill ships in, so that order reflects how popular those repositories are rather than any review of Harness Step 3 Session Management against them. Open each page to compare what they document and how they install.

More from simbajigege/book2skills

Harness Step 3 Session Management is one of 25 skills cataloged on DirSkills from simbajigege/book2skills.

See all 25 skills
🧠
1h ago

Agent Memory Implementation

Agent Memory Implementation restructures a crowded MEMORY.md into a small pointer index plus on-demand topic files. Use it when MEMORY.md gets too long, mixes pointers with content, or needs stale memories corrected in place.
AI Engineering
16130
🛠️
1h ago

Agent Tool Builder

Agent Tool Builder helps define agent tools with a fail-closed pattern that keeps schema, security metadata, and execution together. Use it when creating a new tool or adding validation and permission checks to an existing one.
AI Engineering
16130
📘
1h ago

Business Adventures Analysis

Business Adventures Analysis uses John Brooks’s *Business Adventures* as a diagnostic lens for business failures, market panics, governance issues, and innovation questions. Use it to map a live situation to recurring patterns like hubris, incentive drift, and institutional fragility.
AI Engineering
16130
📈
1h ago

Common Sense Index Investing

Common Sense Index Investing applies Bogle's rules to fund selection, fees, asset allocation, and buy-and-hold discipline. Use it to compare index vs active funds, assess cost drag, and simplify a portfolio.
Data
16130
🧠
1h ago

Compact Memory Implementation

Compact Memory Implementation shows how to add memory compaction to an agent by triggering a forked compactor, generating a structured summary, and restoring it in later sessions. Use it when building context compression or persistence with Claude Agent SDK or the Anthropic API.
AI Engineering
16130
📣
1h ago

Contagious Viral Content Berger

Contagious Viral Content Berger applies Jonah Berger's STEPPS framework to diagnose why content spreads or stalls. Use it to engineer shareability with social currency, triggers, emotion, public visibility, practical value, and stories.
AI Engineering
16130