← 返回 Skills 市场
whoisme007

Memory Integration

作者 whoisme007 · GitHub ↗ · v0.1.0 · MIT-0
cross-platform ⚠ suspicious
218
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install memory-integration
功能描述
同步 OpenClaw 记忆文件到共现图和语义向量库,支持增量更新及语义加共现的统一搜索接口。
安全使用建议
This package likely implements the claimed memory-sync feature, but several inconsistencies make it untrustworthy until verified. Before installing or enabling: 1) Request the full source or repository (homepage/source unknown). 2) Confirm presence and contents of the required adapter modules (integration.adapter.co_occurrence_adapter and integration.adapter.semantic_vector_adapter) and inspect them for network I/O or credential usage. 3) Verify which environment variable controls the workspace/config path in your runtime (the code uses OPENCLAW_WORKSPACE; SKILL.md documents different vars/defaults) and ensure the config path is confined to a safe workspace. 4) Test in an isolated environment (non-production agent) to observe file reads/writes and any outbound connections. 5) Ask the author to fix naming/path mismatches (MemoryIntegrationAdapter vs MemoryIntegration, memory-integration vs memory-sync-enhanced) or provide a reconciled release; if the author cannot explain these inconsistencies, treat the skill as suspicious and avoid enabling autonomous invocation. If you can obtain and review the missing adapter modules and they look benign (no unexpected network endpoints or secrets exfiltration), the concerns would be largely resolved and this could be reclassified as benign.
功能分析
Type: OpenClaw Skill Name: memory-integration Version: 0.1.0 The memory-integration skill is a legitimate utility designed to synchronize OpenClaw memory files with co-occurrence and semantic vector stores. The core logic in scripts/memory_integration.py handles file parsing, MD5-based change detection, and integration with local adapters for enhanced search capabilities. No evidence of data exfiltration, unauthorized command execution, or malicious prompt injection was found; the script operates strictly within the expected workspace environment.
能力评估
Purpose & Capability
The SKILL.md describes memory synchronization to a co-occurrence engine and a semantic vector store; the included Python implements parsing memory files, computing IDs, updating co-occurrence, and optional vector-store usage — broadly aligned with the stated purpose. However, naming mismatches (SKILL.md refers to MemoryIntegrationAdapter while the code provides MemoryIntegration) and inconsistent default paths (SKILL.md: ~/workspace/... vs code default /root/.openclaw/workspace) reduce confidence that the package is well-formed.
Instruction Scope
SKILL.md and example usage instruct importing MemoryIntegrationAdapter from integration.adapter.memory_integration_adapter, but the repository only contains scripts/memory_integration.py with a MemoryIntegration class. The runtime code appends a hard-coded absolute sys.path (/root/.openclaw/workspace/skills/memory-sync-enhanced/scripts) that references a different name ('memory-sync-enhanced'), and expects external adapter modules (integration.adapter.co_occurrence_adapter, integration.adapter.semantic_vector_adapter) that are not included. These inconsistencies mean an agent following SKILL.md may fail or load adapters from unexpected locations; the instructions also allow reading and writing workspace memory files and a sync config file (within the workspace) which is consistent with purpose but should be confirmed.
Install Mechanism
No install spec is provided (instruction-only with included script). That minimizes install-time risk but means runtime behavior depends on environment layout and presence of adapter plugins. Nothing in the package attempts to download or execute remote installers.
Credentials
SKILL.md declares environment variables (MEMORY_SYNC_CONFIG, SYNC_INTERVAL_HOURS, ENABLE_SEMANTIC_SYNC, ENABLE_COOCCURRENCE_SYNC) but the code only reads OPENCLAW_WORKSPACE (default /root/.openclaw/workspace) and does not reference MEMORY_SYNC_CONFIG or the other declared vars. This mismatch is suspicious: either the documentation is stale or the code relies on implicit workspace paths. The code writes/reads a sync config under workspace/integration/memory_sync_config.json — this is reasonable for a memory-sync tool but you should verify the effective path and that no unrelated files are accessed. No credentials are requested.
Persistence & Privilege
The skill is not always:true and is user-invocable (normal). It writes a sync state file inside the workspace and manipulates files under workspace/memory and MEMORY.md; it does not request system-wide privileges or modify other skills' configs in the supplied code. Autonomous invocation is allowed by default (not a separate concern) but combine with other red flags before enabling fully autonomous usage.
如何使用
  1. 确保已安装 OpenClaw(本地或 Docker 部署)
  2. 在对话框中输入安装命令:/install memory-integration
  3. 安装完成后,直接呼叫该 Skill 的名称或使用 /memory-integration 触发
  4. 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
v0.1.0
memory-integration v0.1.0 - 初始发布:为 OpenClaw 星型记忆架构提供记忆同步与增强搜索服务 - 支持原生记忆结构(MEMORY.md, memory/*.md)同步至共现图和语义向量库 - 提供统一的语义+共现增强搜索 API - 实现增量同步与唯一记忆 ID,支持跨系统引用 - 提供标准适配器接口与配置支持
元数据
Slug memory-integration
版本 0.1.0
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 Integration 是什么?

同步 OpenClaw 记忆文件到共现图和语义向量库,支持增量更新及语义加共现的统一搜索接口。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 218 次。

如何安装 Memory Integration?

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

Memory Integration 是免费的吗?

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

Memory Integration 支持哪些平台?

Memory Integration 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 Integration?

由 whoisme007(@whoisme007)开发并维护,当前版本 v0.1.0。

💬 留言讨论