dsh-knowledge-graph
Run the following command in DeepSeek Harness:
dsh plugin install cwbcheng/dsh-knowledge-graph
Paste the following prompt into your AI chat to install this plugin:
To install this plugin in DeepSeek Harness, run dsh plugin install cwbcheng/dsh-knowledge-graph. Then use the Agent to execute cordis_define and cordis_run to load the plugin. Source: https://github.com/cwbcheng/dsh-knowledge-graph
About this plugin
When dealing with long-form books, technical docs, or massive chat logs, information is often fragmented and hard to link. The dsh-knowledge-graph plugin for DeepSeek Harness aims to transform dry text or images into structured knowledge graphs. Leveraging the AI's powerful decomposition capabilities, it automatically breaks down text into 8 node types and 12 relationship types, and even supports generating knowledge graphs directly from images. What’s more, its bidirectional positioning feature—clicking a node to jump to the source text and vice versa—completely solves the pain point of "graph vs. text separation."
Screenshots
Use Cases
- Decompose long technical documents or books into structured knowledge graphs.
- Clicking graph nodes jumps directly to source text, achieving bidirectional linking.
- Visualize AI conversation execution traces to analyze reasoning and evidence sources.
Best For
- Researchers needing to organize large amounts of literature and academic notes.
- Technical writers focused on document understanding and knowledge association.
- Managers wishing to convert unstructured conversations into structured knowledge.
Related Plugins
Traceable, searchable cross-session memory for AI agents that turns conversation knowledge into a typed knowledge graph and recalls relevant subgraphs instead of replaying full history, natively integrated with DeepSeek Harness.
Gives DSH AI cross-session long-term memory, to-do and skill management, plus multi-session orchestration, external AI delegation, and an infinite canvas that grows with you.
dsh-mnemon is a three-tier, pluggable, Agent-driven memory system for DeepSeek Harness, combining Runtime memory, Project Documents, and replaceable Memory Spaces with nine long-term providers.
An opinionated, zero-infrastructure file-based memory protocol: human-readable markdown files plus a bounded always-loaded index, curation discipline, and no database or embeddings.