← 返回 Skills 市场

Memory Vault

Name: Memory Vault
Author: ndp

作者 Andy Peterson · GitHub ↗ · v1.0.0 · MIT-0

cross-platform ⚠ suspicious

总下载

当前安装

版本数

在 OpenClaw 中安装

/install ndpsoftware-memory-vault

功能描述

Enables autonomous agents to store, retrieve, and recall long-term, durable memory fragments and state across independent sessions.

使用说明 (SKILL.md)

NDP Memory Vault Skill

This skill connects autonomous bots (eg. OpenClaw) to the persistent cloud storage backend.

Programmatic Capabilities

Agents can use this endpoint to append logs, retain vector fragments, and recall state variables across independent, headless environment lifecycles.

Intended Use

Store durable notes and execution state.
Retrieve prior context between sessions.
Keep lightweight memory fragments for long-running agents.
Persist agent preferences, project notes, and task state.
Cache reusable context for headless or multi-step workflows.
Record compact summaries that can be recalled later.

Search Keywords

OpenClaw, bot memory, agent memory, persistent memory, long-term context, session memory, durable memory, knowledge vault, state recall, notes, reminders, storage, vector fragments, session state, persistent storage, headless agent, autonomous bot

安全使用建议

Review before installing. This is not shown to be malicious, but only use it if you trust the Memory Vault provider and are comfortable with agents saving and later reusing persistent cloud memory. Set rules to avoid storing secrets or personal data, require approval for sensitive memory writes, and confirm there is a way to review and delete stored memory.

能力评估

ℹ Purpose & Capability

The stated purpose is coherent with the capability: SKILL.md says it will "Store durable notes and execution state" and "Retrieve prior context between sessions." This is expected for a memory skill, but it is inherently sensitive.

⚠ Instruction Scope

SKILL.md says agents can "append logs, retain vector fragments, and recall state variables across independent, headless environment lifecycles," but it does not define what may be stored, when user approval is needed, or how recalled memory should be trusted.

ℹ Install Mechanism

There is no install spec or code to review, while SKILL.md declares an external "endpoint_url" and "auth_type: bearer_token." This is not malicious by itself, but the registry metadata does not clearly declare the credential contract.

⚠ Credentials

The skill connects agents to a "persistent cloud storage backend" for "project notes" and "task state," but the artifacts do not describe storage boundaries, retention, deletion, export, privacy limits, or per-project separation.

⚠ Persistence & Privilege

The core behavior is durable cross-session memory: SKILL.md advertises "long-term context," "persistent storage," and state recall for "long-running agents." That can materially affect future agent behavior if memory is stale, sensitive, or poisoned.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install ndpsoftware-memory-vault
安装完成后，直接呼叫该 Skill 的名称或使用 /ndpsoftware-memory-vault 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.0

Initial release of NDP Memory Vault skill. - Enables autonomous bots to connect to persistent cloud storage. - Supports log appending, vector fragment retention, and state variable recall. - Facilitates durable notes and execution state storage across sessions. - Designed for headless agents and long-running workflows needing persistent memory.

元数据

Slug ndpsoftware-memory-vault

版本 1.0.0

许可证 MIT-0

累计安装 0

当前安装数 0

历史版本数 1

常见问题

Memory Vault 是什么？

Enables autonomous agents to store, retrieve, and recall long-term, durable memory fragments and state across independent sessions. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 63 次。

如何安装 Memory Vault？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install ndpsoftware-memory-vault」即可一键安装，无需额外配置。

Memory Vault 是免费的吗？

是的，Memory Vault 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。

Memory Vault 支持哪些平台？

Memory Vault 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 Memory Vault？

由 Andy Peterson（@ndp）开发并维护，当前版本 v1.0.0。