JohnPitter

Imprint — AI skill for Claude Code

AI community

Persistent memory for AI coding agents — Claude Code plugin with Go + Svelte.

How to install Imprint

This entry records only its repository, not the path inside it, so there is no exact command to give. Open JohnPitter/imprint and copy the folder into ~/.claude/skills/, or the file into ~/.claude/agents/.

What Imprint does

Persistent memory for AI coding agents — Claude Code plugin with Go + Svelte.

Alternatives in AI

Osaurus — Own your AI. The native macOS harness for AI agents -- any model, persistent memory, autonomous execution, cry 5.1k ★
Agentmemory — #1 Persistent memory for AI coding agents based on real-world benchmarks 1.9k ★
Nocturne Memory — A lightweight, rollbackable, and visual Long-Term Memory Server for MCP Agents 1.3k ★

README

**Persistent memory for AI coding agents — every session builds on the last.**

[![Go](https://img.shields.io/badge/Go-1.25+-00ADD8?style=flat-square&logo=go&logoColor=white)](https://go.dev) [![Svelte](https://img.shields.io/badge/Svelte-5-FF3E00?style=flat-square&logo=svelte&logoColor=white)](https://svelte.dev) [![SQLite](https://img.shields.io/badge/SQLite-WAL-003B57?style=flat-square&logo=sqlite&logoColor=white)](https://sqlite.org) [![CI](https://img.shields.io/github/actions/workflow/status/JohnPitter/imprint/ci.yml?branch=master&style=flat-square&label=CI)](https://github.com/JohnPitter/imprint/actions) [![License](https://img.shields.io/badge/License-MIT-orange?style=flat-square)](#license)

[Features](#features) · [How It Works](#how-it-works) · [Install](#install) · [Architecture](docs/ARCHITECTURE.md) · [Changelog](CHANGELOG.md) · [Development](#development)

What is Imprint?

Imprint is a **plugin for Claude Code and Codex** that gives your AI agent persistent memory across sessions. Every tool call, decision, and discovery is captured, compressed via a cheap LLM, indexed for search, and injected as context into future sessions.

Since **v2.0** it is also a **token-economy + high-value-memory** plugin: it measures the token *saldo* (context saved − background LLM spent), organizes memory into **three explicit layers** (base → refined → rooted "intuitions"), injects lazily, and runs the background work on a cheap model — **Haiku** on Claude Code, **GPT-5 mini/nano** on Codex.

**No Docker. No external databases. Single Go binary + SQLite.**

Works the same on both agents. On Claude Code it auto-detects the Claude OAuth token (Haiku); on Codex it reuses your `codex login` (ChatGPT OAuth) or an `OPENAI_API_KEY` (GPT-5). When an Anthropic key is present it takes priority; the providers are a fallback chain, so both can coexist in one install.

Inspired by [age