Your AI should remember
what you built yesterday.

Context Engineering Kit stores project state in files. AI reads them every session. No more cold starts.

$ git clone https://github.com/Vincent-chao-lang/context-engineering-kit /tmp/cek && /tmp/cek/install.sh
Works with Claude Code. Adapts to Codex, Cursor, Windsurf.

The difference is 15 minutes per session.

Before every session

You: This project is a VAT system using Tauri...
You: We use Zhipu API for OCR and translation...
You: Last week we decided against SQLite...
You: There's a bug in PDF parsing...
You: We're working on batch recognition...
AI: Got it. What do you need?
You: (finally start working, 15 min later)

After /start

AI: Read CLAUDE.md, DECISIONS.md, memory/...
AI:
AI: Phase: MVP development
AI: Last: batch invoice recognition (30 min ago)
AI: Bug:  PDF parsing garbled on some files
AI: Next: fix PDF issue, then add progress bar
AI:
AI: Ready. What should I work on?

It's not just about saving time.

The real cost of AI amnesia isn't minutes. It's wrong decisions, repeated bugs, and broken continuity.

Continuity

Every session picks up where the last one ended. Progress, blockers, next steps — all preserved.

Before: "What was I doing last Friday?"
After:  /start reads memory/current_state.md
         Last: auth refactor (2 days ago)
         Next: write tests for login flow

Decision consistency

Recorded decisions don't get relitigated. AI reads why you rejected a proposal and never suggests it again.

Before: AI: "Let's use MongoDB!"
         You: "We discussed this last week..."
After:  You: "Check DECISIONS.md D001"
         AI: "Right, rejected MongoDB for SQLite."
🛡

Bug awareness

Known bugs stay visible. AI won't reintroduce a fix or ignore a known issue when writing new code.

Before: AI fixes bug, next session
         re-introduces the same broken pattern
After:  /start shows active bugs
         AI avoids known-broken patterns
         /wrap moves fixed bugs to resolved

Core files. Three commands.

The kit installs project docs, memory files, prompts, and Claude Code commands. The files hold context; the commands read and maintain it.

1

CLAUDE.md + AGENTS.md — tool entry docs

Claude Code reads CLAUDE.md. Codex and other repo-instruction tools can read AGENTS.md. Both point AI at the same project context.

2

ARCHITECTURE.md + DECISIONS.md + TASKS.md — shared facts

Architecture, decisions, and current tasks live in markdown. AI can see what exists, why choices were made, and what should happen next.

3

memory/ + prompts/ — working state and rules

Progress, bugs, daily logs, lessons learned, and coding rules stay outside chat history. /start reads them; /wrap updates them.

Is this for you?

If any of these sound familiar, yes.

Side project

"I only code 1-2 hours a day"

Every minute counts. You can't afford to spend 15 minutes re-explaining the project to AI each session. /start gives it all back in 30 seconds.

Small team

"We step on each other's toes"

Team mode shows who's working on what. Personal memory is gitignored, shared decisions use merge=union. No merge conflicts on state files.

Open source

"I come back to this project once a month"

Context doesn't expire. Open the project, run /start, and AI knows where you left off — even if it was weeks ago.

Learning

"I follow tutorials and lose track"

AI remembers which tutorial step you're on, what worked, what didn't. Your learning progress becomes part of the project context.

Common questions

How is this different from Claude Code's built-in memory?
Claude Code memory (.claude/memory/) stores user preferences and general background — like "I prefer TypeScript" or "use conventional commits". This kit stores project state: what's done, what's broken, what was decided and why, what's next. They complement each other. Built-in memory is about who you are; this kit is about where the project is.
Does it pollute my git repo?
CLAUDE.md, AGENTS.md, DECISIONS.md, TASKS.md, and ARCHITECTURE.md are committed — they're useful documentation for humans too. In team mode, personal memory/ files are gitignored. Everything is plain markdown, no binary or generated code.
Can I use it with Cursor, Codex, or Windsurf?
Yes. The core idea is tool-agnostic: project facts live in files. AGENTS.md gives Codex a repo-level entry point. Cursor and Windsurf can use the same facts through their rules systems; see the cross-tool section below.
Can I customize the templates?
Everything is markdown. Edit any file, add sections, remove what you don't need. The templates are starting points, not constraints. Commands under .claude/commands/ are plain markdown too — change the steps to match your workflow.
Solo or team mode?
Start with solo (default). If you need team collaboration, run install.sh --team --user yourname. It creates shared memory, personal memory, memory/TEAM.md, and git merge rules. Existing files are not deleted; move useful solo memory into the team structure when you establish the team baseline.
Can I use it on an existing project?
Yes. /init-context analyzes your existing codebase and generates context files. If you already have CLAUDE.md, it reads and preserves it, only adding missing sections.

Try it in 60 seconds.

Install step 1
# In your project root
git clone https://github.com/Vincent-chao-lang/context-engineering-kit /tmp/cek
/tmp/cek/install.sh
Initialize step 2
# In Claude Code
/init-context

# AI analyzes your project,
   generates CLAUDE.md, AGENTS.md,
   ARCHITECTURE.md, TASKS.md,
   DECISIONS.md, memory/...

# Review and commit
git add . && git commit
Daily loop step 3
# Start of day
/start
# AI reads context, ready to work

# ... code, commit, code ...

# End of day
/wrap
# AI updates memory for next time

Works with your favorite tool.

Same files. Different entry points.

Claude Code
CLAUDE.md
Codex
AGENTS.md
Cursor
.cursor/rules
Windsurf
rules

Read the playbooks.

Books and demo material for learning, sharing, and launching the methodology.

Mini handbook

Mini Handbook

Run AI project memory in one afternoon.

Complete book

Complete Practice Book

A 15-chapter path from tool usage to team habits.

Demo script

Demo Script

Show before/after install, /start, /wrap, and team mode.

Used by developers who are tired of re-explaining their projects.

stars on GitHub
Star on GitHub