NxcoreAI/NxMem

No repository description was reported by the source.

Open original source ↗

Older snapshot: check the original source for current information.

From the publisher

README & documentation

Source preview

Read the project’s overview, installation instructions and usage examples. The original README is the source of truth.

仿人类记忆模型的 Agent 上下文引擎——为 LLM Agent 提供可持久化、可检索、可推理的长期记忆。 [](https://www.typescriptlang.org/) [](https://fastify.dev/) [](https://react.dev/) [](https://neo4j.com/) [](https://pnpm.io/) [](https://www.gnu.org/licenses/agpl-3.0) Nexcore Context Engine 把「短期记忆 → 长期记忆」的认知过程落实为一条可观测、可评测的流水线:将对话与文档摄入为事件(Event),经 LLM 抽取为结构化事实(Fact),做时间轴融合后按策略准入短期记忆(STM),再通过「做梦」式巩固(Dreaming / Consolidation)沉淀为长期记忆(LTM),最终在需要时完成检索与 Context Pack 组装,为下游 Agent 提供带来源引用的上下文。 引擎借鉴人类记忆的「短时 → 长时」结构,把记忆生命周期拆成若干可独立观测的层: 做梦巩固的七维评分定义、分数锚点与校准样例见 STM 到 LTM 七维评分 Prompt。 启动后打开 http://localhost:5173 即可进入调试台,写入测试记忆事件并查看解析、事实、STM、LTM 与融合时间轴。 默认配置文件为 config/context-engine.json,后端与前端代理都会读取。可用 CONTEXTENGINECONFIG 指向另一份配置: 配置结构(示例): 部署或本地调试时,以下环境变量会覆盖配置文件: longMemEval.graphStore.mode 默认为 inherit(沿用主 graphStore);设为 local 只用独立 SQLite;设为 neo4j 使用 longMemEval.graphStore.neo4j,未填写的字段继承主 graphStore.neo4j。 后端扫描固定目录 data/inbox 的第一层文件(.txt / .md / .json / .docx),生成 MemoryEvent 进入解析链路。.docx 从 word/document.xml 提取正文(暂不提取批注、修订、页眉页脚);.xlsx / .pptx / 图片会生成 unsupported 事件。 去重规则:file path + mtime + size 生成幂等 key,同一文件重复扫描不会重复写入,文件修改后视为新版本重新摄入。 对 序号 / 事实发生时间 / 记忆类型 / 事实记忆 / 来源类型 测试表,用确定性脚本导入(不调用 LLM): 脚本默认读取根目录的 极核产品经理一周事实记忆假数据.md,写入 synthetic-test 租户,并可用 --file / --dataset-id / --tenant / --principal 覆盖。 响应使用 memory-graph.v1,节点与边共用顶层 page 但各自管理 limit,可用…

NxcoreAI/NxMem on GitHub A short preview, not the full document.

Read the full README ↗ · Preview checked 2026-10-31T18:52:24.441Z

Inside the original README — Document outline
  1. 🧠 Nexcore Context Engine
  2. 目录
  3. ✨ 核心特性
  4. 🧠 记忆模型
  5. 🏗️ 架构
  6. 🛠️ 技术栈
  7. 📁 项目结构
  8. 🚀 快速开始
  9. 运行前提
  10. 安装与启动
  11. 一条链路跑通
  12. ⚙️ 配置

Headings are captured from the source. Links open the publisher’s document, not a locally hosted copy.

Links from the README

References supplied by the publisher, not independently verified endorsements. Check the destination before downloading files or entering credentials.

Documentation belongs to its respective authors. Reported project/model license: Other. A listing is not a grant of reuse or training rights. Confirm the document’s own terms at the source.

What this repository does

The source does not provide a repository description. Read the project README before evaluating it.

Repository facts

Owner
NxcoreAI
Primary language
TypeScript
Stars
1,087
Forks
107
Open issues + pull requests
0
License
Other
Archived
No
Default branch
main
Created
2026-09-10T12:14:47.000Z
Last push
2026-09-22T13:20:01.000Z

Topics and intended use

No topics were included in the latest source metadata.

Review the README for scope, installation, examples and limitations. We do not run repository code or certify it.

Evaluate before installing

Review licensing and dependencies, inspect recent commits and unresolved issues, and test in an isolated environment before production use. Stars and forks alone cannot answer these questions.

README and project files

Issues and maintenance discussion

Releases and changelog

Source and freshness

Source: GitHub. Metadata observed 2026-10-01T12:05:27.301Z. Daily imports are snapshots, not real-time monitoring.

Popularity and source listings do not establish security, suitability, licensing rights or benchmark performance.

Open original source

Related repositories

freeCodeCamp/freeCodeCamp

morluto/rea

openclaw/openclaw

nilbuild/developer-roadmap

deepseek-ai/deepseek-harness

anomalyco/opencode