Agent Skills › TanStack/ai › tanstack-ai-memory-in-memory

tanstack-ai-memory-in-memory

GitHub

提供基于内存Map的零依赖记忆适配器，适用于本地开发、单元测试及单进程演示。不支持持久化或多进程部署，适合轻量级场景，超出容量建议切换Redis。

packages/ai-memory/skills/tanstack-ai-memory-in-memory/SKILL.md TanStack/ai

Trigger Scenarios

需要快速搭建本地开发环境的记忆功能编写单元测试或集成测试时模拟记忆存储构建无需数据持久化的单进程演示Demo

Install

npx skills add TanStack/ai --skill tanstack-ai-memory-in-memory -g -y

More Options

Non-standard path

npx skills add https://github.com/TanStack/ai/tree/main/packages/ai-memory/skills/tanstack-ai-memory-in-memory -g -y

Use without installing

npx skills use TanStack/ai@tanstack-ai-memory-in-memory

指定 Agent (Claude Code)

npx skills add TanStack/ai --skill tanstack-ai-memory-in-memory -a claude-code -g -y

安装 repo 全部 skill

npx skills add TanStack/ai --all -g -y

预览 repo 内 skill

npx skills add TanStack/ai --list

SKILL.md

Frontmatter

{
    "name": "tanstack-ai-memory-in-memory",
    "description": "Use when wiring inMemory() from @tanstack\/ai-memory\/in-memory — explains setup, options (embedder, extract, topK\/minScore), when to pick it (dev\/tests\/single-process demos), and what NOT to use it for (multi-process or persistent)."
}

In-Memory Memory Adapter

Zero-dependency recall/save adapter backed by a Map. Records vanish on process restart.

When to use it

Local development.
Vitest / Playwright tests.
Single-process demos where users don't need persistence.

When NOT to use it

Production multi-process deployments — every worker has its own Map; users get inconsistent memory.
Anything that needs survival across restarts.

For production, use redis() (see the tanstack-ai-memory-redis skill).

Setup

import { memoryMiddleware } from '@tanstack/ai-memory'
import { inMemory } from '@tanstack/ai-memory/in-memory'

const memory = inMemory()

memoryMiddleware({ adapter: memory, scope })

Options

inMemory(options?) accepts:

topK (default 6), minScore (default 0.15), kinds — recall tuning.
embedder: { embed(text): Promise<number[]> } — enable semantic scoring (both recall and save embed through it).
extract(turn, scope) — return derived facts to persist alongside the raw turn (e.g. call an LLM to pull out preferences). Without it, save stores the raw user/assistant messages and recall scores them lexically + by recency.
render(hits) — replace the built-in prompt renderer.