research

Verify factual claims via web search with evidence hierarchy and source credibility checks.

110|14|Updated May 12, 2026
One-click install
npx skills add https://github.com/EliasOulkadi/shokunin --skill research-eliasoulkadi
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: research
Source: https://github.com/EliasOulkadi/shokunin/tree/main/.pack/skills/research
Command: npx skills add https://github.com/EliasOulkadi/shokunin --skill research-eliasoulkadi

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents hallucinations and fake citations by verifying claims with a strict evidence hierarchy before any writing happens.

Core Features & Use Cases

  • Source verification & triangulation: Confirms each fact by prioritizing primary sources and cross-checking across independent references.
  • Evidence hierarchy with publication rules: Classifies sources into weighted levels and enforces using only verified/corroborated information for professional outputs.
  • Fact-checking workflows for document pipelines: Runs before downstream steps like drafting/humanization to ensure citations, dates, versions, names, and statistics are traceable.
  • Error handling for common research failures: Detects single-source gaps, unverifiable claims, paywalls, and inconsistencies, then marks or omits accordingly.

Quick Start

Ask the AI to research the topic and return only VERIFIED or CORROBORATED claims with evidence hierarchy guidance and citations.

Frequently Asked Questions about research

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

FAQPage Schema
How do I fact-check claims and prevent AI hallucinations before writing?

Fact-checking claims before writing involves verifying details against primary sources using an evidence hierarchy. This process cross-checks independent references to classify information validity, ensuring only verified or corroborated facts are published to prevent hallucinations.

What is source triangulation for web research?

Source triangulation in web research confirms a fact by cross-checking it across multiple independent references. This technique prioritizes primary sources over secondary ones to validate claims and ensure details are traceable before drafting documents.

How do I verify citations and prevent fake references in generated documents?

Verifying citations requires classifying sources by evidence level and validating dates, names, and statistics against reputable references. Enforcing publication rules ensures only verified or corroborated claims with traceable citations are used in document pipelines.

Can I use evidence hierarchy rules to filter out unverifiable claims?

Yes, applying an evidence hierarchy filters out unverifiable claims by classifying sources into weighted levels. This method detects single-source gaps and inconsistencies, omitting or marking unverified information to prevent unverified publication.

What is the best way to handle research failures like paywalls or single-source gaps?

Handling research failures like paywalls or single-source gaps requires error detection during source verification. The system identifies missing data and inconsistencies, then marks or omits the unverifiable claims accordingly to maintain research integrity.

Does fact-checking work for deep topic investigation and security version verification?

Fact-checking works for deep topic investigation and security version verification by validating specific details against primary references. This ensures accurate version numbers and statistics are corroborated before drafting or humanizing final outputs.