← 返回 Skills 市场
V19 Code Memory Triplet Store
作者
Liuyanfeng1234
· GitHub ↗
· v1.0.0
· MIT-0
51
总下载
0
收藏
0
当前安装
1
版本数
在 OpenClaw 中安装
/install v19-code-memory-triplet-store
功能描述
结构化存储"代码片段↔关键知识点↔发生情境"的三元组记忆,实现代码与知识的直接关联检索。
使用说明 (SKILL.md)
V19 Code-Memory Triplet Store v1.0.0
结构化存储"代码片段 ↔ 关键知识点 ↔ 发生情境"的三元组记忆,实现代码与知识的直接关联检索。
核心概念
传统的代码检索只能根据文件名或关键词搜索。V19三元组记忆将每个代码片段与它承载的知识点和发生的业务情境绑定:
代码片段 ──关联── 关键知识点
│ │
└────发生情境───────┘
三元组结构
{
"code_snippet": {
"file": "trust_score_api.py",
"function": "calculate_trust_score",
"lines": "42-78"
},
"knowledge_point": {
"domain": "V19治理/信任分",
"concept": "信任分四维度加权计算",
"key_insight": "认知运营型Agent的审计通过率权重高于行为执行型"
},
"context": {
"scenario": "Agent认证冲刺",
"trigger": "墨言达到60分基础认证门槛",
"outcome": "V19-CERT-001授予"
}
}
检索能力
按代码查知识
输入: trust_score_api.py
输出: 信任分计算公式、VPAV等效机制、认证门槛分布
按知识查代码
输入: "VPAV验证一次等效多少次普通调用"
输出: vpav_check.py:verify_procedure()、trust_score_api.py:L85
按情境查代码和知识
输入: "Agent首次自助注册时发生了什么"
输出: governance/register.py + Agent接入流程文档 + Gemma4-Local-Agent案例
与知识拓扑审计的集成
三元组存储在V19知识拓扑中扮演关键角色:
- 认知熵检测:如果某个代码片段被关联了过多不相关的知识点 → 认知熵过高
- 知识孤岛扫描:如果某个关键知识点未被任何代码片段引用 → 知识孤岛
- 代码-知识一致性:VPAV验证 Skill Manifest 声明与实际代码实现的一致性
调用示例
# 检索三元组
curl -s https://boat-atlas-spa-flexible.trycloudflare.com/governance/knowledge \
-H "X-Governance-Key: \x3C你的专属密钥>" \
-H "Content-Type: application/json" \
-d '{"query":"trust_score","mode":"triplet"}'
公开体验
公开密钥: v19-e5d585e28439decc614f09f91c4caa8c
curl -s https://boat-atlas-spa-flexible.trycloudflare.com/governance/health \
-H "X-Governance-Key: v19-e5d585e28439decc614f09f91c4caa8c"
自助注册
curl -s -X POST https://boat-atlas-spa-flexible.trycloudflare.com/governance/register \
-H "Content-Type: application/json" \
-d '{"agent_name":"你的Agent名称"}'
信任锚点
安全使用建议
This skill appears safe to install as documentation-only, but use its curl examples carefully. Verify the V19 endpoint and operator, keep private governance keys secret, and avoid sending sensitive code or internal project details unless you trust the service.
功能分析
Type: OpenClaw Skill
Name: v19-code-memory-triplet-store
Version: 1.0.0
The skill instructs the AI agent to interact with an external, ephemeral endpoint (boat-atlas-spa-flexible.trycloudflare.com) for 'governance' and 'registration' purposes. The use of a Cloudflare Tunnel URL in SKILL.md to transmit agent metadata and queries is a high-risk behavior often associated with data collection or tracking, and lacks the transparency expected for a production-grade knowledge management tool.
能力评估
Purpose & Capability
The stated purpose and SKILL.md content consistently describe a code-knowledge-context triplet memory and retrieval system. The actual functionality appears to depend on an external API rather than local code.
Instruction Scope
The commands are presented as user-facing examples, not automatic execution. If followed, they send queries or registration data to an external service, so use should remain user-directed.
Install Mechanism
There is no install specification, no binaries, and no code files. The static scanner had no runnable code to analyze and reported no suspicious patterns.
Credentials
The skill does not request local files, OS access, or environment variables, but it does reference an external HTTPS endpoint and an X-Governance-Key header.
Persistence & Privilege
The skill discusses persistent triplet storage in a V19 knowledge topology, but the artifacts do not show local persistence, background execution, or elevated 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 v19-code-memory-triplet-store - 安装完成后,直接呼叫该 Skill 的名称或使用
/v19-code-memory-triplet-store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
v1.0.0: 代码-知识-情境三元组结构化存储,实现跨域关联检索。
元数据
常见问题
V19 Code Memory Triplet Store 是什么?
结构化存储"代码片段↔关键知识点↔发生情境"的三元组记忆,实现代码与知识的直接关联检索。 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件,目前累计下载 51 次。
如何安装 V19 Code Memory Triplet Store?
在 OpenClaw 或 Claude Code 对话框中运行命令「/install v19-code-memory-triplet-store」即可一键安装,无需额外配置。
V19 Code Memory Triplet Store 是免费的吗?
是的,V19 Code Memory Triplet Store 完全免费,采用 MIT-0 许可证,可自由下载、安装和使用。
V19 Code Memory Triplet Store 支持哪些平台?
V19 Code Memory Triplet Store 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 V19 Code Memory Triplet Store?
由 Liuyanfeng1234(@liuyanfeng1234)开发并维护,当前版本 v1.0.0。
推荐 Skills