← 返回 Skills 市场
Human-Like Memory Plugin
作者
HumanLikeTeam
· GitHub ↗
· v1.0.1
· MIT-0
160
总下载
0
收藏
0
当前安装
7
版本数
在 OpenClaw 中安装
/install humanlike-memory-plugin
功能描述
Long-term memory plugin for OpenClaw: automatic recall, storage, and agent tools
使用说明 (SKILL.md)
Human-Like Memory Plugin
Long-term memory plugin for OpenClaw. Gives your agent the ability to remember past conversations, user preferences, and important context across sessions.
Features
- Automatic memory recall before each response
- Automatic conversation storage after each response
- Agent-callable tools:
memory_searchandmemory_store - Registered as a first-class memory slot (
kind: "memory") - Runtime reads OpenClaw plugin config only
- Privacy-preserving by default: platform metadata extraction is disabled unless explicitly enabled
- Requires OpenClaw >= 2026.2.0
Setup
1. Get API Key
Visit plugin.human-like.me → Register → Copy your mp_xxx key
2. Install
openclaw plugins install @humanlikememory/human-like-mem
3. Configure
# Set API key (required)
openclaw config set plugins.entries.human-like-mem.config.apiKey "mp_your_key_here"
# Set as the active memory engine
openclaw config set plugins.slots.memory human-like-mem
# Enable memory search for the agent
openclaw config set agents.defaults.memorySearch '{"enabled":true}' --strict-json
4. Restart
openclaw restart
5. Verify
openclaw status
You should see:
Memory │ 0 files · 0 chunks · sources remote-api · plugin human-like-mem · vector ready
Configuration Options
All options via openclaw config set plugins.entries.human-like-mem.config.\x3Ckey> \x3Cvalue>:
| Key | Type | Default | Description |
|---|---|---|---|
apiKey |
string | (required) | Your API key from plugin.human-like.me |
baseUrl |
string | https://plugin.human-like.me |
API endpoint |
userId |
string | openclaw-user |
User identifier |
agentId |
string | auto-detected | Agent identifier |
recallEnabled |
boolean | true |
Enable automatic memory recall |
addEnabled |
boolean | true |
Enable automatic memory storage |
memoryLimitNumber |
number | 6 |
Max memories to recall per turn |
minScore |
number | 0.1 |
Minimum relevance score (0-1) |
minTurnsToStore |
number | 5 |
Store after N conversation turns |
stripPlatformMetadata |
boolean | true |
Do not send Feishu/Discord platform IDs unless explicitly enabled |
Agent Tools
Once installed, your agent can actively use memory:
memory_search— Search past conversations and stored knowledgememory_store— Save important information for future recall
Troubleshooting
| Problem | Solution |
|---|---|
| "API key not configured" | openclaw config set plugins.entries.human-like-mem.config.apiKey "mp_xxx" |
| "unavailable" in status | Set plugins.slots.memory to human-like-mem |
| Check logs | openclaw logs | grep "Memory Plugin" |
License
Apache-2.0
安全使用建议
This plugin appears to be what it claims: a remote memory backend that will send conversation text (and optional platform IDs) to https://plugin.human-like.me. Before installing: 1) Understand privacy impact — the default configuration auto‑stores and retains memories indefinitely unless you delete them; avoid sending passwords, API keys, or sensitive PII. 2) Start with a test API key and set addEnabled: false and/or minTurnsToStore high while you evaluate. 3) Keep stripPlatformMetadata: true (default) unless you explicitly need platform IDs. 4) Because the package includes runnable code, verify the package source (Git repo or npm) and confirm the baseUrl matches the vendor; you can also inspect network traffic (mitmproxy) during testing as suggested in SECURITY.md. 5) Note the metadata/README minor inconsistencies (small doc defaults differ); ask the maintainer for the canonical source repository and changelog if you need higher assurance.
功能分析
Type: OpenClaw Skill
Name: humanlike-memory-plugin
Version: 1.0.1
The plugin implements a long-term memory system by storing and retrieving conversation history via a remote API (plugin.human-like.me). Its behavior is transparent and strictly aligned with its stated purpose, featuring privacy-preserving defaults such as the automatic stripping of platform-specific metadata (Feishu/Discord IDs) in index.js. While it uses prompt-injected instructions to guide the AI agent's handling of retrieved memories, these are standard for Retrieval-Augmented Generation (RAG) and are well-documented in the provided SECURITY.md and PRIVACY.md files.
能力评估
Purpose & Capability
Name/description, config schema, README/SECURITY/PRIVACY, and included code all indicate a memory plugin that sends conversation content to an external memory API and exposes memory_search/memory_store tools. Requiring a plugin config API key (plugins.entries.human-like-mem.config.apiKey) is proportionate to that purpose.
Instruction Scope
SKILL.md instructs installing the plugin, setting the plugin config API key, enabling the memory slot and agent memorySearch — everything is within the scope of a memory plugin. It explicitly documents what is sent and how to disable auto-storage or platform metadata. Note: defaults enable auto-storage and auto-recall (addEnabled/recalIEnabled true), which means conversation content will be sent by default; this is a behavioral/privacy choice worth highlighting.
Install Mechanism
Registry metadata claims 'No install spec — instruction-only', but the package includes executable plugin code (index.js and plugin.js) and an openclaw.plugin.json extension entry. This is likely a packaging/metadata mismatch rather than malicious but should be clarified: the plugin contains code that will run in the host, not purely textual instructions.
Credentials
No environment variables or unrelated credentials are requested; the only required secret is the plugin API key stored in OpenClaw config (plugins.entries.human-like-mem.config.apiKey). The code explicitly parses platform IDs (Feishu/Discord) but documents that such extraction is disabled by default (stripPlatformMetadata = true).
Persistence & Privilege
The skill is not forced-always, does not request system-wide config changes beyond its own plugin config, and does not request elevated privileges. It will run as an OpenClaw memory plugin at the documented lifecycle hooks (before_prompt_build, agent_end, session_end). Autonomous invocation is normal for plugins; nothing here elevates that privilege.
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 humanlike-memory-plugin - 安装完成后，直接呼叫该 Skill 的名称或使用
/humanlike-memory-plugin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
v1.0.1
Normalize wrapped recall queries.
v1.0.0
Release 1.0.0: stable automatic memory plugin with ClawHub-safe manifest alignment, config-only credential handling, privacy-first defaults, and preserved automatic recall/store hooks.
v0.4.9
Harden manifest and runtime for ClawHub scan clarity; keep automatic memory hooks, require apiKey in config schema, default to stripped platform metadata, remove upgrade-notification path, and document config-only setup.
v0.4.8
Add env var documentation and mappings for recallEnabled, addEnabled, stripPlatformMetadata
v0.4.6
增强安全透明度：添加 SECURITY.md 和 PRIVACY.md，新增 stripPlatformMetadata 配置选项
v0.4.5
移除自引用依赖，添加 SECURITY.md 安全说明文档
v0.4.4
长期记忆插件：自动记忆召回、存储，支持 memory_search 和 memory_store 工具
元数据
常见问题
Human-Like Memory Plugin 是什么？
Long-term memory plugin for OpenClaw: automatic recall, storage, and agent tools. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 160 次。
如何安装 Human-Like Memory Plugin？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install humanlike-memory-plugin」即可一键安装，无需额外配置。
Human-Like Memory Plugin 是免费的吗？
是的，Human-Like Memory Plugin 完全免费，采用 MIT-0 许可证，可自由下载、安装和使用。
Human-Like Memory Plugin 支持哪些平台？
Human-Like Memory Plugin 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 Human-Like Memory Plugin？
由 HumanLikeTeam（@humanlike2026）开发并维护，当前版本 v1.0.1。
推荐 Skills