thomasog/coldstart

Get any folder ready for AI agents, once.
The next agent picks up where the last one stopped, on any machine.
An illustration of a typical run, not a screen recording.
## Why
No lock-in. Your instructions and memory don't belong to one agent. Switch agents and keep working.
It all lives in the folder. Nothing important is trapped in an account or in an app's hidden data.
It travels. Copy, sync or zip the folder, and the context goes with it.
It lasts. Plain text you can still read years from now, with no special tool.
## What it does
Run coldstart in a code repo or in a personal project such as a move, a trip or paperwork. It lists the folder, asks a few questions, writes the files below, and stops. It doesn't open your documents or start the project's work; that belongs to the next session.
├── AGENTS.md # instructions any agent can read
├── .ai/
│ ├── HANDOFF.md # where we stopped and what comes next
│ └── memory/
│ └── MEMORY.md # index of durable facts
├── .gitignore # only if you want git
└── .claude/ # only if you use Claude Code
├── CLAUDE.md # imports AGENTS.md, the handoff and the memory index
├── settings.json # keeps plans and memory inside the folder
└── agents/
├── scout.md # fast and cheap: search, read, bulk vision
└── advisor.md # strong second opinion for hard calls
The habit that makes it work: agents read .ai/HANDOFF.md at the start of a session and rewrite it at the end.
## Works with
| Agent | Status |
| Claude Code | Tested |
| Codex | Tested |
| Cursor, Copilot, Antigravity | Read AGENTS.md according to their documentation; not yet run |
## Install
Paste this into your agent:
text
Install the agent skill in skills/coldstart from https://github.com/thomasog/coldstart, then tell me how to run it.
Or use a command:
bash
npx skills add thomasog/coldstart
bash
gh skill install thomasog/coldstart coldstart --agent codex --scope user
Claude Code plugin:
bash
claude plugin marketplace add thomasog/coldstart
bash
claude plugin install coldstart@thomasog-plugins
By hand: copy skills/coldstart/ to ~/.agents/skills/, or to ~/.claude/skills/ for Claude Code.
## Use
Ask for it by name: /coldstart in Claude Code, $coldstart in Codex, or "run coldstart" anywhere. It never starts on its own. If the folder has no git, start your agent from the folder itself.
## What it reads, writes and runs
Reads: file names, agent instruction files such as CLAUDE.md, and project manifests such as package.json. Nothing outside the folder.
Writes: the files above, and nothing else. It moves old instruction files to .ai/scratch/ instead of deleting them, and shows you any command, URL or install step they contain first.
Runs: skills/coldstart/scripts/coldstart.py, a short Python script that uses only the standard library, makes no network calls and asks no questions. Without Python, the agent does the same steps by hand.
Changes in Claude Code: .claude/settings.json turns off auto memory for that folder, so memory stays in .ai/memory/. Anything Claude had already saved for the folder stays in ~/.claude and is no longer loaded there; copy what you want by hand. The same file blocks reading .env files.