skills.rest
Shallow-W
Shallow-WCOMMUNITY
Explore all skills by this author →
GitHub

research-wiki

Maintains a persistent project knowledge base of papers, experiments, and claims with typed relationships.

Updated Apr 21, 2026
Search Tags:#context retrieval#knowledge base#claims#experiments#graph relations#research-wiki#literature ingestion
One-click install
npx skills add https://github.com/Shallow-W/llm-wiki --skill research-wiki-shallow-w
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: research-wiki
Source: https://github.com/Shallow-W/llm-wiki/tree/main/.claude/skills/research-wiki
Command: npx skills add https://github.com/Shallow-W/llm-wiki --skill research-wiki-shallow-w

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It turns scattered papers, experiments, and conclusions into a persistent research knowledge base so you do not have to re-derive everything from scratch each time you iterate.

Core Features & Use Cases

  • Persistent research memory (papers, ideas, experiments, claims): Store structured pages with stable node IDs and an append-only log.
  • Typed relationship graph with auto-generated connections: Track evidence-bearing relationships in graph/edges.jsonl and keep Connections sections consistent via generation.
  • Fast, budgeted retrieval via query_pack: Produce an 8k-char compressed summary optimized for model context windows, prioritizing top gaps and failed ideas.

Quick Start

Run the wiki initialization for the current project by calling: research-wiki init.

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 base from scattered papers and experiments?▼

To build a persistent research knowledge base, ingest your papers, ideas, experiments, and claims into structured pages with stable node IDs. This maintains an append-only log so you do not re-derive context from scratch across iterations.

What is the best way to track typed relationships and evidence graphs across research literature?▼

Tracking evidence-bearing relationships requires storing typed connections in a graph edges file as the single source of truth. Auto-generated Connections sections keep your research relationship graph consistent without manual edge updates.

How do I initialize a research wiki for my current project?▼

Initializing a research wiki requires running the wiki initialization command for your current project. This sets up the canonical helper script and creates the graph edges file needed to store relationship data.

Can I generate compressed context summaries prioritizing research gaps for model context windows?▼

Fast context retrieval generates an 8k-char compressed summary optimized for model context windows. This budgeted query pack prioritizes top research gaps and failed ideas for downstream workflows.

Do I need a specific helper script to maintain a research knowledge base?▼

Maintaining this research knowledge base requires the presence of a canonical helper script. It relies on this file alongside a graph edges log to execute synchronization backfills, graph updates, and health checks.

Related Skills

Explore More Education & Research Skills→
sjtuytcsjtuytc
27

research-wiki

Organize papers, ideas, experiments, and claims into a structured knowledge base.

Community
Advanced
zhuyingqinzhuyingqin
1

research-wiki

Maintain a per-project research knowledge base with typed relationships.

Community
Advanced
SaschaBrunnerCHSaschaBrunnerCH
19

arcgis-knowledge-graphs

Query and visualize ArcGIS Knowledge graphs using openCypher.

Community
Advanced
gaos6egaos6e

ontology

Create and validate typed knowledge graphs of entities and relations.

Community
Advanced