Warmboot Full — AI skill for Claude Code

Warmboot Full

Warmboot Full gives AI coding agents memory and rules across sessions. Without shared state, a new session can start from zero: agents forget decisions, redo finished work, trust stale results and lose each other's changes. Warmboot Full keeps a contract and the task's state as plain text in the repository, read through AGENTS.md or CLAUDE.md. This checkout also holds guidance for kickoff, maintenance, upgrades and releases, plus structural acceptance checks and cold-start exercises for the protocol.

Designed and directed by Tiago Fonseca; implemented and reviewed with coding agents under this protocol. This is the full edition of Warmboot; [Which edition?](#which-edition) compares it with [Warmboot Light](https://github.com/korreio/warmboot).

**Status: experimental.** Warmboot is a collection of practices for running coding agents. It makes no promise of better results. Use it at your own risk and take what is useful.

Mermaid source

flowchart TD
    accTitle: Agents lose context; Warmboot installs nine files; each session reads, works, proves and hands off
    P["Agents lose scope, overwrite work, reuse stale results"] --> I["init installs nine plain-text files"]
    I --> S
    subgraph S["Each session, then the next"]
        direction LR
        R["Read the task and the core"] --> W["Work inside the named scope"]
        W --> V["Prove: record what ran, on which inputs"]
        V --> H["Hand off: the next session checks those inputs"]
    end

Try it

From a clone of this repository (replace the demo path with your project's absolute path to install for real):

mkdir -p .acceptance-work/readme-demo