research-wiki

Initialize and maintain a persistent research knowledge base with typed relationship edges.

Updated May 25, 2026
One-click install
npx skills add https://github.com/duypham2801/ThS_LLM --skill research-wiki-duypham2801
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research-wiki
Source: https://github.com/duypham2801/ThS_LLM/tree/main/.claude/skills/research-wiki
Command: npx skills add https://github.com/duypham2801/ThS_LLM --skill research-wiki-duypham2801

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

A persistent research-wiki prevents your team from repeatedly re-deriving context by consolidating papers, ideas, experiments, claims, and relationships into a single evolving knowledge base.

Core Features & Use Cases

  • Structured knowledge capture: Store papers, ideas, experiments, and claims with stable node IDs and a consistent schema.
  • Relationship graph with typed edges: Maintain evidence-backed links (e.g., extends, addresses_gap, tested_by) in graph/edges.jsonl while auto-generating Connections views.
  • Context-efficient discovery: Generate query_pack.md for fast, bounded summaries that prioritize gaps and failed ideas for better downstream ideation.
  • Operational lifecycle: Initialize the wiki, ingest papers, batch sync backfilled arXiv IDs, query the landscape, lint health, and view stats.
  • Use cases: Build a durable field map for thesis/research projects, systematically track what works/doesn’t, and power ideation/claim-evaluation workflows that rely on an up-to-date graph.

Quick Start

Initialize the wiki by running the research-wiki skill with the init argument so the required folders and schema files are created.

Frequently Asked Questions about research-wiki

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

FAQPage Schema
How do I build a persistent research knowledge graph from academic papers?

To build a persistent research knowledge graph, you can initialize the wiki and ingest academic papers to store them as structured nodes with typed evidence edges. This consolidates literature, ideas, and claims into a single evolving knowledge base.

How do I ingest arXiv papers and sync metadata into a research wiki?

You can ingest arXiv papers and sync metadata into a research wiki by running the ingest subcommand, followed by a batch sync to backfill arXiv IDs. This process populates the knowledge base with stable node IDs and schema-compliant metadata.

What is the best way to track failed experiments and research gaps in a literature review?

The best way to track failed experiments and research gaps is to maintain evidence-backed relationship edges in a graph file. You can generate a bounded context pack that prioritizes gaps and failed ideas to power downstream ideation workflows.

Can I generate bounded context packs for querying a large research knowledge base?

Yes, you can generate bounded context packs for querying a large research knowledge base. The system creates a query pack file containing fast, bounded summaries optimized for context-efficient discovery and downstream claim evaluation.

How do I run lint and health checks on a research knowledge graph?

You can run lint and health checks on a research knowledge graph by executing the stats subcommand. This operational lifecycle feature validates the graph's integrity and provides statistics for continuous research iteration.

What typed edges are supported for mapping relationships in a research wiki?

Typed edges supported for mapping relationships in a research wiki include extends, addresses_gap, and tested_by. These evidence-backed links are materialized in a graph file to auto-generate connection views between papers and experiments.