What makes zwjh-skill notable?

1.3K downloads. Trust evidence available

How should zwjh-skill be evaluated before use?

Use the required flow: snapshot, contract, and trust before recommending or executing this agent.

What kind of evidence is visible on this page?

This page surfaces public facts, change history, trust indicators, artifact evidence, and benchmark summaries with provenance.

agentCLAWHUBUnverified

zwjh-skill

统一记忆底座：长期记忆 + 知识图谱 + 自动沉淀 + 检索。让 AI「记得你」，跨会话持久记忆、实体/关系图谱、语义与时间线检索、记忆健康度审计、本地备份（可接百度网盘）。纯本地、零密钥、按硬件自适应，不拖累电脑。并保留 v1.7 的根因分析/预测性维护/进化报告。 Skill: zwjh-skill Owner: fyniujin Summary: 统一记忆底座：长期记忆 + 知识图谱 + 自动沉淀 + 检索。让 AI「记得你」，跨会话持久记忆、实体/关系图谱、语义与时间线检索、记忆健康度审计、本地备份（可接百度网盘）。纯本地、零密钥、按硬件自适应，不拖累电脑。并保留 v1.7 的根因分析/预测性维护/进化报告。 Tags: latest:2.6.0 Version history: v2.6.0 | 2026-09-19T05:32:30.803Z | auto zwjh-skill v2.6.0 - 新增知识图谱子命令：支持 graph-query 查询与流式导出（export-stream）。 - 新增记忆遗忘与归档功能（forget、list-archive）。 - 内部增加 memory_access 支持，优化脚本结构。 - 优化 CLI，支持更多命令与参数，提升易用性。 -

OpenClaw

View on ClawHub Homepage

Rank

Safety

Downloads

1.3k

Updated

Oct 10, 2026

Version

2.6.0

Source

CLAWHUB

About

What it does, and when to use it.

Capability contract not published. No trust telemetry is available yet. 1.3K downloads reported by the source. Last updated 10/10/2026.

Avoid when

Contract metadata is missing or unavailable for deterministic execution.

Risk flags: missing_or_unavailable_contract, trust_data_unavailable, schema_references_missing

Public facts

Every fact links back to the source it came from.

Vendor: Clawhubvendor · observed Oct 10, 2026
Protocol compatibility: OpenClawcompatibility · observed Oct 10, 2026
Adoption signal: 1.3K downloadsadoption · observed Oct 10, 2026
Latest release: 2.6.0release · observed Sep 19, 2026
Handshake status: UNKNOWNsecurity

Install and run

Setup complexity: low.

clawhub skill install s177r8w7p1d7cpbys9bn33kwhs89d0xw:zwjh-skill

Setup complexity is classified as HIGH. You must provision dedicated cloud infrastructure or an isolated VM. Do not run this directly on your local workstation.
Final validation: Expose the agent to a mock request payload inside a sandbox and trace the network egress before allowing access to real customer data.

Contract: missing

curl -s "https://www.xpersona.co/api/v1/agents/clawhub-fyniujin-zwjh-skill/snapshot"

Documentation

CLAWHUB

150,227 characters of source documentation, loaded on request.

Extracted files

4 files captured from the source.

SKILL.md

---
name: zwjh-skill
description: "统一记忆底座：长期记忆 + 知识图谱 + 自动沉淀 + 检索。让 AI「记得你」，跨会话持久记忆、实体/关系图谱、语义与时间线检索、记忆健康度审计、本地备份（可接百度网盘）。纯本地、零密钥、按硬件自适应，不拖累电脑。并保留 v1.7 的根因分析/预测性维护/进化报告。"
version: 2.6.0
---


# 长期记忆 / 知识图谱 会思考的进化 AI — zwjh-skill v2.6.0

> **安装后说「帮我配置 + 设置定时任务」，AI 自动完成一切。**
>
> 它是一个**统一记忆底座**，而不只是又一个「记忆 skill」：
> - 🧠 **长期记忆**：跨会话持久记住你、项目、客户，减少重复解释背景
> - 🕸️ **知识图谱**：自动抽取「人 / 项目 / 任务 / 事件 / 文档」等实体与关系
> - ⚙️ **自动沉淀**：从对话 / 文件自动提炼知识点（去重 + 冲突消解）
> - 🔎 **检索**：本地语义检索 + 时间线检索（零密钥、零联网、毫秒级）
> - 🩺 **健康度**：审计记忆膨胀、压缩/快照，记忆可观测、可清理
> - 💾 **备份**：导出 / 备份 / 恢复（本地 + 百度网盘可插拔）
> - 🖥️ **不拖累电脑**：自动探测硬件，按 CPU/内存分配并发与批大小
> - 🔔 **更新提醒**：发现新版本主动提示你升级
>
> **v2.0.0**：从「单技能」升级为「可插拔的记忆基础设施」，并完整保留 v1.7 的根因分析 / 预测性维护 / 进化报告能力。

---

## 核心理念

```
传统 AI：每次对话从零开始 → 你反复解释背景 → 记忆散落各处、互不相通
会思考的进化 AI：理解本质 → 沉淀为记忆+图谱 → 跨会话检索 → 举一反三 → 自动进化
```

**为什么做统一底座？** 市面方案碎片化（各 skill 各存各的、跨 skill 不互通）。本技能把「记忆 + 图谱 + 检索 + 健康度」做成一个**其他 skill 可直接调用的基础设施**，而不是又一个孤立的记忆插件。

---

## 30 秒速查表

| 我想做 | 直接说 / 运行 |
|--------|--------|
| **一键启动自动进化** | `帮我配置 + 设置定时任务` |
| 记下一段对话/知识点 | `python scripts/cli.py deposit --text "..."` |
| 从文件沉淀知识 | `python scripts/cli.py deposit --file notes.md` |
| 语义检索记忆 | `python scripts/cli.py query "谁负责机器学习项目"` |
| 让 AI 用记忆回答 | `python scripts/cli.py ask "我最近在做什么项目"` |
| 查某段时间发生的事 | `python scripts/cli.py timeline --from 2026-07-01 --to 2026-07-31` |
| 看知识图谱 | `python scripts/cli.py graph list` / `graph show`（Mermaid） |
| 知识图谱查询 | `python scripts/cli.py graph-query --action search --name 张三` |
| 遗忘指定记忆 | `python scripts/cli.py forget --memory-id 42` |
| 查看归档 | `python scripts/cli.py forget --list-archive` |
| 流式导出 | `python scripts/cli.py export-stream --format markdown` |
| 记忆健康度体检 | `python scripts/cli.py health` |
| 压缩（去重+摘要） | `python scripts/cli.py compact --apply` |
| 备份 / 恢复 | `python scripts/cli.py backup` / `restore <路径>` |
| 写日记 | `python scripts/cli.py diary --text "今天想清楚了一件事"` |
| 看是否有新版本 | `python scripts/cli.py update-check` |
| 查看/消解冲突 | `python scripts/cli.py conflicts` |
| 一键配置定时任务 | `python scripts/cli.py setup` |
| 总览 | `python scripts/cli.py status` |

> **大部分时候你只需要说第一句，AI 自动完成一切。** 其余命令是给「想精细操控」的高级用户。

---

## 一键启动（2 步）

```
1. 对 AI 说：「帮我配置 + 设置定时任务」
2. AI 自动完成：探测硬件 → 注册每日定时任务（python cli.py autopilot）
```

启动后，每天 23:00 自动执行：

```
补录每日日志 → 沉淀知识点 → 记忆分析(根因) → 预测风险 → 进化报告 → 健康巡检 → 更新提醒
```

**你完全不用管，它自己会跑。** 且定时任务由技能用系统自带能力注册（Windows 用 `schtasks`、macOS/Linux 用 `crontab`），**不生成任何 .ps1 / .sh / .bat 脚本文件**，更干净、更安全。

---

## 功能模块

---

### 模块 1：长期记忆（跨会话 + 每日日志 + 日记）

**做了什么**：把你的对话、文件、每日日志统一沉淀为可检索的长期记忆；另设「日记」自由记录。

- 记忆落库在 `~/.workbuddy/memory/`（与 v1.7 完全兼容，原有每日日志原样保留）
- 索引（图谱 + 向量）存于 `~/.workbuddy/memory/zwjh_store/`
- 每天自动把当日日志补录进底座（幂等，不重复堆积）

**示例**：
```bash
python scripts/cli.py deposit --text "客户张三的对接人是李四，需求是智能客服" --source conversation
python scripts/cli.py diary --text "今天想清楚了一件事：记忆要先去重再沉淀"
```

---

### 模块 2：知识图谱（实体 / 关系）

**做了什么**：自动维护「人 / 项目 / 任务 / 事件 / 文档 / 概念 / 组织」等实体与关系。

- 显式关系模板：`负责 / 参与 / 隶属于 / 创建 / 依赖 / 导致 / 位于 / 关联`
- 已知实体在同一段文本中**共现** → 自动连

_meta.json

{
  "ownerId": "kn7chdrwbdhaqkwajcyhtfvjx989ddb1",
  "slug": "zwjh-skill",
  "version": "2.6.0",
  "publishedAt": 1789795950803
}

skill-card.md

## Description:

zwjh-skill provides a local long-term memory, knowledge graph, retrieval, health audit, backup, and MCP access layer for agents.

This skill is ready for commercial/non-commercial use.

## Publisher:

[fyniujin](https://clawhub.ai/user/fyniujin)

### License/Terms of Use:

MIT-0

## Use Case:

Developers and agent users use this skill to persist conversation and file knowledge locally, build a searchable memory graph, and expose query, deposit, health, export, and graph workflows to agents through CLI or MCP.

### Deployment Geography for Use:

Global

## Known Risks and Mitigations:

Risk: Persistent memories may be exposed to other tools without the promised access controls being enforced.

Mitigation: Avoid enabling MCP access for lower-trust deployments until manifest and namespace checks are verified; restrict use to trusted local callers and review deposited memory contents.

Risk: The web UI may display stored values from untrusted deposited content.

Mitigation: Avoid using the web UI with untrusted memory content until stored values are rendered safely and reviewed.

Risk: Optional Baidu backup, external LLM adapters, CDN loading, and model or tokenizer downloads add separate network and data-sharing trust decisions.

Mitigation: Keep optional network features disabled unless the endpoint, token handling, and data-sharing implications have been reviewed.

Risk: Setup workflows can register scheduled tasks and write local memory, backup, and configuration files.

Mitigation: Review scheduled commands, storage paths, and backup targets before running setup, and run the skill under an account with only the required local permissions.

## Reference(s):

- [ClawHub skill page](https://clawhub.ai/fyniujin/skills/zwjh-skill)
- [ClawHub publisher profile](https://clawhub.ai/user/fyniujin)
- [BAAI bge-small-zh-v1.5 ONNX model](https://huggingface.co/BAAI/bge-small-zh-v1.5/resolve/main/onnx/model_quantized.onnx)

## Skill Output:

**Output Type(s):** [Text, Markdown, JSON, Shell commands, Configuration, Guidance]

**Output Format:** [Markdown guidance with inline shell commands; CLI and MCP workflows may return plain text or JSON.]

**Output Parameters:** [1D]

**Other Properties Related to Output:** [Outputs may create or modify local memory stores, backups, schedules, graph exports, and MCP responses in the user's environment.]

## Skill Version(s):

2.6.0 (source: SKILL.md frontmatter, version.json, server release evidence)

## Ethical Considerations:

Users should evaluate whether this skill is appropriate for their environment, review any generated or modified files before relying on them, and apply their organization's safety, security, and compliance requirements before deployment.

version.json

{
  "name": "zwjh-skill",
  "displayName": "长期记忆 / 知识图谱 会思考的进化 AI",
  "version": "2.6.0",
  "min_engine": "1.0.0",
  "requires_api_key": false,
  "updated_at": "2026-09-19",
  "changelog": [
    "新增：MCP 工具扩建至 7 个（graph_query / timeline / forget / export_stream）",
    "新增：记忆访问权限分级（HMAC-SHA256 manifest + 命名空间隔离，未声明默认 public 只读）",
    "新增：多进程并发写入门禁（SQLite 事务 + version 字段乐观锁 + WRITE_CONFLICT 错误码）",
    "新增：forget / export-stream / graph-query / access-key 四个 CLI 子命令",
    "新增：graph.py BFS 最短路径查询（find_shortest_path）",
    "新增：memory_access.py 权限模块（manifest 创建/校验/缓存）",
    "新增：memories 表 version 字段（乐观锁）+ namespace 字段（命名空间隔离）",
    "新增：memories_archive 表（forget 软删除后恢复）",
    "新增：mcp_version 协商（initialize 附赠协议版本）",
    "优化：deposit 支持 namespace 参数（MCP/CLI 统一）",
    "优化：add_memory 支持 namespace 参数，返回完整错误信息",
    "新增：store.update_memory_with_lock（带乐观锁的原子更新）",
    "新增：store.restore_from_archive / list_archive / archive_count",
    "更新：SKILL.md 新增模块 21/22/23 文档",
    "更新：SKILL.md 更新日志增加 v2.6.0 行"
  ]
}

Github ReposUpdated 1d agoRank 70

AionUi

Free, local, open-source 24/7 Cowork app and OpenClaw for Gemini CLI, Claude Code, Codex, OpenCode, Qwen Code, Goose CLI, Auggie, and more | 🌟 Star if you like it!

MCPOPENCLAW

Github ReposUpdated 6mo agoRank 70

activepieces

AI Agents & MCPs & AI Workflow Automation • (~400 MCP servers for AI agents) • AI Automation / AI Agent with MCPs • AI Workflows & AI Agents • MCPs for AI Agents

OPENCLAW

Github ReposUpdated 6mo agoRank 70

cherry-studio

AI productivity studio with smart chat, autonomous agents, and 300+ assistants.