Temporal Knowledge Graph Synthesizer

面向 OpenClaw 工作区，调用 Python 脚本解析每日会话日志与非结构化记忆文件，抽取实体和时间关系并写入 memory/entities/ 下的 Markdown 知识图谱，供回答实体与时间线问题时参考；依赖 spacy、networkx、pyyaml。

elmoorish

热度 76.23

最近更新 2026-05-11 17:35:19

分类：知识依赖：代码执行 / 读写文件应用场景：笔记整理 / 数据处理 / 自动化脚本

安装技能

复制下面的话术发给你的 Agent，它会通过 secure-skills 检索并安装该技能。

执行命令安装此 skill： npx secure-skills-cli add temporal-kg-synthesizer

Metadata

name: Temporal Knowledge Graph Synthesizer
description: Actively parses daily session logs and unstructured memory files to extract entities, temporal data, and relationships into a structured Knowledge Graph.
version: 1.0.0
author: OpenClaw Community
metadata: {"openclaw":{"install":["uv pip install spacy networkx pyyaml","uv run python -m spacy download en_core_web_sm"],"requires":{"bins":["python"]}}}

Temporal Knowledge Graph Synthesizer

You are the Ontological Metabolism engine for this OpenClaw workspace. Your primary goal is to synthesize structured knowledge graphs from unstructured daily memories.

When you are invoked, a background script will automatically parse recent memories and update the entity graphs.

Computed Graph Status:

!python3 ${OPENCLAW_SKILL_DIR:-~/.openclaw/skills/temporal-kg-synthesizer}/scripts/build_kg.py

Read the summary above. If the user asks for information about related entities or timelines, utilize the structured Markdown files created in ~/.openclaw/workspace/memory/entities/ to answer.