skills.rest
Immortalqx
ImmortalqxCOMMUNITY
Explore all skills by this author ā†’
WebsiteGitHub

research-wiki

Builds and maintains a persistent per-project research knowledge base of papers, ideas, experiments, and claims.

52|Updated Apr 24, 2026
Search Tags:#knowledge graph#knowledge base#literature review#arxiv#research management#paper ingestion#research wiki
One-click install
npx skills add https://github.com/Immortalqx/my_codex_skills --skill research-wiki-immortalqx
Or copy as Structured Prompt for Agentā–¼
Please help me install this Agent Skill.
Skill: research-wiki
Source: https://github.com/Immortalqx/my_codex_skills/tree/main/research-wiki
Command: npx skills add https://github.com/Immortalqx/my_codex_skills --skill research-wiki-immortalqx

SYSTEM DOCUMENTATION & REQUIREMENTS

šŸ’” This Skill includes scripts (resource) components.

What problem does it solve? Research projects lose context between sessions: papers get re-read, failed ideas get retried, and relationships between prior work get re-derived from scratch. This Skill maintains a persistent, structured wiki that accumulates papers, ideas, experiments, and claims with typed relationships, so project knowledge compounds instead of evaporating. ## Core Features & Use Cases - Structured Paper Ingest: Fetch arXiv metadata automatically and generate schema-consistent paper pages with deduplication, slug generation, and index rebuilds. - Typed Relationship Graph: Record edges like extends, contradicts, supports, and invalidates in a JSONL graph connecting papers, ideas, experiments, and claims. - Compact Query Pack: Generate a budget-capped (8000 char) context file summarizing gaps, failed ideas, key papers, and relationship chains for fast project re-entry. - Use Case: While surveying a new research direction, ingest each arXiv paper you read, log failed experiment ideas, then run the query command before your next session to reload a compressed field map instead of re-reading everything. ## Quick Start Initialize a research wiki for this project and ingest the arXiv paper 1706.03762 with a one-line thesis.

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 for a project?ā–¼

Initialize a wiki directory with the init subcommand, then ingest papers by arXiv ID or manual metadata. Each paper becomes a structured Markdown page, and relationships are recorded as typed edges in a JSONL graph file.

How to ingest arXiv papers into a local wiki with Python?ā–¼

Run the bundled research_wiki.py script with the ingest_paper command and an --arxiv-id flag. It fetches metadata from the arXiv Atom API, generates a slug, deduplicates against existing pages, and rebuilds the index and query pack.

Can I add papers without an arXiv ID?ā–¼

Yes, manual ingest is supported by passing --title, --authors, and --year instead of --arxiv-id. This fallback also triggers automatically if the arXiv API fetch fails and manual metadata was supplied.

Does the research wiki track failed research ideas?ā–¼

Yes, ideas are a first-class entity type with outcome fields, and failed ideas are prioritized in the generated query pack. Recording negative outcomes explicitly prevents repeating dead-end approaches in later sessions.

What are the limitations of the arXiv metadata fetch?ā–¼

The fetch depends on network access to the arXiv API and raises an error on timeouts or malformed responses. Non-arXiv venues require manual metadata entry, and the helper only supports single-paper or batch ID-based ingestion.

Related Skills

Explore More Education & Research Skillsā†’
Lianggs8Lianggs8
12

arxiv

Search arXiv by keyword or ID and download paper PDFs.

Community
Intermediate
OpenLAIROpenLAIR
1.1k

aris-research-wiki

Maintains a persistent research knowledge base of papers, ideas, experiments, and claims.

Official
Advanced
lix965996-artlix965996-art

novelty-check

Extract three core claims and compare them against recent arXiv literature.

Community
Intermediate
PsychQuantPsychQuant

che-zotero-mcp

Manage Zotero libraries, search academic literature, and generate citations.

Official
Advanced