Verified:GitHub

dalu-dongguan

Extract entities and compute evidence-backed semantic, co-occurrence, temporal, source-consistency, and causal links from text.

10|1|Updated May 4, 2026

Search Tags:#knowledge graph #causal inference #semantic similarity #entity extraction #temporal analysis #hallucination risk #link detection

Ready to run

npx skills add https://github.com/isLinXu/under-one --skill dalu-dongguan

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: dalu-dongguan
Source: https://github.com/isLinXu/under-one/tree/main/underone/skills/dalu-dongguan
Command: npx skills add https://github.com/isLinXu/under-one --skill dalu-dongguan

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill finds semantic, co-occurrence, temporal, source-consistency, and causal relationships across multiple text segments so you can surface evidence-backed insights instead of guessing.

Core Features & Use Cases

Multi-type link detection: extracts entities and builds a relation graph covering semantic similarity, entity co-occurrence, temporal markers, source consistency, and causal markers.
Evidence-oriented outputs: produces link reports, anomaly signals, and a hallucination-risk score to help agents verify and ask better follow-ups.
Visualization-ready: generates Mermaid code for quickly viewing the resulting knowledge graph and link structure.

Quick Start

Ask the agent to run the link detector on your segments list (segments.json) to produce link_report.json with links, entity_map, and mermaid_code.

Frequently Asked Questions about dalu-dongguan

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I build a knowledge graph from unstructured text segments?▼

You can detect cross-segment relationships by extracting entities and computing semantic, co-occurrence, temporal, source-consistency, and causal links. This process generates a structured JSON link report and Mermaid graph code for visual analysis.

What is the best way to find hidden causal links across multiple text segments?▼

Finding hidden causal links across text segments requires analyzing causal markers and entity co-occurrence to map relationships. This approach outputs structured temporal and causal chains alongside anomaly signals to reveal evidence-backed insights.

How do I verify narrative claims and detect anomalies across multiple documents?▼

Verifying narrative claims involves computing source-consistency and semantic links across segment sets to identify contradictions. This yields anomaly signals and a hallucination-risk score to help agents ask better follow-up questions.

Can I generate Mermaid graph code automatically from extracted entity relationships?▼

You can generate Mermaid graph code automatically from extracted entity relationships by running a link detector on your segments list. The process outputs visualization-ready code representing the detected semantic and causal link structure.

Does link detection from text segments output a hallucination risk score?▼

Link detection from text segments outputs a hallucination risk score alongside structured JSON containing detected links and entity mapping. This helps agents verify multi-source correlations and assess the reliability of extracted insights.