← 返回 Skills 市场
894
总下载
0
收藏
6
当前安装
1
版本数
在 OpenClaw 中安装
/install agent-memory-1-0-0
功能描述
Persistent memory system for AI agents to remember facts, learn from experiences, recall memories, and track entities across sessions.
使用说明 (SKILL.md)
AgentMemory Skill
Persistent memory system for AI agents. Remember facts, learn from experience, and track entities across sessions.
Installation
clawdhub install agent-memory
Usage
from src.memory import AgentMemory
mem = AgentMemory()
# Remember facts
mem.remember("Important information", tags=["category"])
# Learn from experience
mem.learn(
action="What was done",
context="situation",
outcome="positive", # or "negative"
insight="What was learned"
)
# Recall memories
facts = mem.recall("search query")
lessons = mem.get_lessons(context="topic")
# Track entities
mem.track_entity("Name", "person", {"role": "engineer"})
When to Use
- Starting a session: Load relevant context from memory
- After conversations: Store important facts
- After failures: Record lessons learned
- Meeting new people/projects: Track as entities
Integration with Clawdbot
Add to your AGENTS.md or HEARTBEAT.md:
## Memory Protocol
On session start:
1. Load recent lessons: `mem.get_lessons(limit=5)`
2. Check entity context for current task
3. Recall relevant facts
On session end:
1. Extract durable facts from conversation
2. Record any lessons learned
3. Update entity information
Database Location
Default: ~/.agent-memory/memory.db
Custom: AgentMemory(db_path="/path/to/memory.db")
安全使用建议
This skill appears coherent for a local agent memory: it stores facts, lessons, and entities in a local SQLite database (default: ~/.agent-memory/memory.db) and requires no external credentials or network access. Before installing: 1) Be aware that memories are stored unencrypted on disk — avoid recording sensitive secrets or consider using a custom db_path with OS-level encryption. 2) Some systems' SQLite builds may lack FTS5 support used for semantic search; test on your environment. 3) Review the included code (src/memory.py) if you want to confirm no additional telemetry or remote calls are added. 4) Because the agent can invoke skills autonomously by default, confirm you trust the skill to store and recall private conversation data.
功能分析
Type: OpenClaw Skill
Name: agent-memory-1-0-0
Version: 1.0.0
The OpenClaw AgentMemory skill bundle is classified as benign. The code provides a persistent memory system using SQLite, with all database interactions appearing to use parameterized queries, mitigating SQL injection risks. There are no external dependencies, no network calls, no arbitrary file system access beyond the designated database file (`~/.agent-memory/memory.db`), and no use of dangerous functions like `os.system`, `subprocess.run(shell=True)`, `eval`, or `exec` with untrusted input. The `SKILL.md` instructions for the AI agent are high-level usage guidelines for the memory API and do not contain any prompt injection attempts or instructions for malicious behavior.
能力评估
Purpose & Capability
Name and description (persistent memory for agents) match the included code and CLI wrappers. The package only requires Python stdlib and SQLite; no unrelated binaries, credentials, or services are requested.
Instruction Scope
SKILL.md limits runtime behavior to creating/reading/updating the local SQLite DB and integrating memory calls into agent lifecycle (session start/end). It does instruct adding memory operations to AGENTS.md/HEARTBEAT.md (expected for a memory system). Nothing in SKILL.md tells the agent to read unrelated system files or contact external endpoints.
Install Mechanism
No install spec is provided (instruction-only in registry), and the bundled source files have no external downloads or package installs. requirements.txt declares no external deps. This is low-risk from an install-execution standpoint.
Credentials
The skill requests no environment variables or credentials. It does create and write a SQLite DB under the user's home (~/.agent-memory/memory.db) by default — expected for a memory store but important to note: data is stored unencrypted on disk and may include sensitive information if the agent stores secrets.
Persistence & Privilege
always is false and the skill does not attempt to modify other skills or global agent configurations. It persists its own state to a local DB which is normal for this purpose.
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 agent-memory-1-0-0 - 安装完成后，直接呼叫该 Skill 的名称或使用
/agent-memory-1-0-0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
AgentMemory 1.0.0 initial release
- Provides a persistent memory system for AI agents to remember facts, learn from experience, and track entities across sessions.
- Offers a simple Python API to store, recall, and learn from information and actions.
- Supports tracking entities with customizable roles and attributes.
- Includes integration guidelines for Clawdbot workflows.
- Uses a local SQLite database for storing memory by default, with configurable location.
元数据
常见问题
Agent Memory 1.0.0 是什么？
Persistent memory system for AI agents to remember facts, learn from experiences, recall memories, and track entities across sessions. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 894 次。
如何安装 Agent Memory 1.0.0？
在 OpenClaw 或 Claude Code 对话框中运行命令「/install agent-memory-1-0-0」即可一键安装，无需额外配置。
Agent Memory 1.0.0 是免费的吗？
是的，Agent Memory 1.0.0 完全免费（开源免费），可自由下载、安装和使用。
Agent Memory 1.0.0 支持哪些平台？
Agent Memory 1.0.0 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 Agent Memory 1.0.0？
由 Sieyer（@sieyer）开发并维护，当前版本 v1.0.0。
推荐 Skills