deep-research-pro

Verify claims against current primary sources with triangulated evidence and confidence scores.

8|1|Updated Mar 30, 2026
One-click install
npx skills add https://github.com/drewid74/ai_skills --skill deep-research-pro-drewid74
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research-pro
Source: https://github.com/drewid74/ai_skills/tree/main/deep-research-pro
Command: npx skills add https://github.com/drewid74/ai_skills --skill deep-research-pro-drewid74

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill prevents hallucinations by forcing every claim to be verified against current, primary sources and by triangulating evidence from multiple independent domains.

Core Features & Use Cases

  • Primary-source focused fact grounding: Prioritizes official docs, repositories, and academic/authoritative materials over secondary summaries.
  • Recency and relevance enforcement: Discards outdated technical data (beyond 18 months) unless it represents a fundamental law or standard.
  • Triangulated confirmation with citations: Marks facts as confirmed only when independently supported by at least two reputable sources.
  • OSINT-style synthesis for live situational awareness: Combines disparate signals (news, reports, and public discussion) into a structured assessment.
  • Best-for tasks: Competitive research, “what changed recently” investigations, specification verification for specific versions, and citation-heavy verification outputs.

Quick Start

Use the deep-research-pro skill to verify the most recent changes to a specific product feature and return only a citation-backed summary with a confidence score.

Frequently Asked Questions about deep-research-pro

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

FAQPage Schema
How do I fact-check technical specs against current primary sources?

To fact-check technical specs, the skill verifies claims by prioritizing official documentation and repositories, discards outdated data beyond 18 months, and confirms facts using two independent reputable sources with a confidence score.

What is the best way to research recent changes for competitive analysis?

For competitive analysis, the skill performs OSINT-style synthesis by expanding queries into sub-queries, gathering signals from at least three distinct domains, and triangulating evidence to produce a structured situational assessment.

How do I prevent AI hallucinations when verifying specific software versions?

Preventing AI hallucinations requires grounding every claim in current web information, enforcing recency validation to discard outdated technical data, and producing a hyperlinked source list for verified spec versions.

Does deep research work with outdated technical data and historical standards?

Deep research discards outdated technical data beyond 18 months unless it represents a fundamental law or standard, ensuring verification focuses on current primary sources while retaining stable historical baselines.

What are the limitations of using OSINT synthesis for citation-based fact-checking?

The limitation of OSINT synthesis is that facts are marked as confirmed only when independently supported by at least two reputable sources, restricting outputs to hyperlinked source lists and confidence scores rather than broad summaries.