deep-reasoning-agent

Verify research drafts through citation validation, fact-checking, and content quality assessment.

1|Updated Feb 15, 2026
One-click install
npx skills add https://github.com/LaZy-Wolf/MAIRA --skill deep-reasoning-agent-lazy-wolf
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-reasoning-agent
Source: https://github.com/LaZy-Wolf/MAIRA/tree/main/backend/skills/deep-reasoning-agent
Command: npx skills add https://github.com/LaZy-Wolf/MAIRA --skill deep-reasoning-agent-lazy-wolf

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill acts as the ultimate quality control for research drafts, ensuring accuracy, completeness, and proper citation before finalization, preventing the dissemination of flawed information.

Core Features & Use Cases

  • Comprehensive Verification: Conducts citation validation, fact-checking, and content quality assessment in a single, structured pass.
  • Automated Scoring: Provides a detailed, weighted score for each verification component.
  • Decision Making: Issues a clear STATUS (VALID, NEEDS_REVISION, INVALID) to guide the research workflow.
  • Use Case: After an AI drafts a research report, this Skill meticulously checks every claim, citation, and piece of content against established criteria, flagging any inaccuracies or omissions.

Quick Start

Use the deep-reasoning-agent skill to verify the research draft provided.

Frequently Asked Questions about deep-reasoning-agent

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

FAQPage Schema
How do I verify research draft accuracy and citations?

To verify research draft accuracy, this Skill performs citation validation, fact-checking, and content quality assessment within a multi-phase workflow, issuing a clear VALID or NEEDS_REVISION status to guide finalization.

What is automated draft verification and how does it work?

Automated draft verification operates within a multi-phase workflow using specialized tools to check claims, citations, and content, providing a detailed weighted score for each verification component to ensure research integrity.

How do I fact-check a research report before finalization?

To fact-check a research report, the Skill meticulously checks every claim and citation against established criteria, flagging inaccuracies or omissions and assigning a clear status of VALID, NEEDS_REVISION, or INVALID.

Can I assess research quality and integrity without manual review?

You can assess research quality without manual review by running the Skill's automated scoring system, which evaluates content completeness and proper citation to prevent the dissemination of flawed information.

Do I need the Tavily API to perform citation validation?

Yes, citation validation requires the Tavily API, along with verification_tools and extract_webpage, to execute the specialized operations needed for comprehensive draft verification.

What are the limitations of automated fact-checking for research drafts?

Automated fact-checking relies on established criteria and web extraction tools, meaning it functions as a quality control layer to flag inaccuracies but still requires human judgment to interpret the final NEEDS_REVISION or INVALID statuses.