deep-research

Orchestrates a 13-agent pipeline for rigorous academic research with APA 7.0 report output.

50|3|Updated Jun 30, 2026
One-click install
npx skills add https://github.com/Dubaoxu/distillation-skills --skill deep-research-dubaoxu
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: deep-research
Source: https://github.com/Dubaoxu/distillation-skills/tree/main/academic-research-skills/deep-research
Command: npx skills add https://github.com/Dubaoxu/distillation-skills --skill deep-research-dubaoxu

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Conducting rigorous academic research requires coordinating question formulation, systematic literature search, source verification, synthesis, and multi-round review — a process that is slow, error-prone, and vulnerable to citation hallucination and confirmation bias when done ad hoc. ## Core Features & Use Cases - 13-Agent Research Pipeline: Coordinates specialized agents across 6 phases (scoping, investigation, analysis, composition, review, revision) with mandatory Devil's Advocate checkpoints and ethics review gates. - 7 Operational Modes: Full research, quick brief, paper review, literature review, fact-check, Socratic guided dialogue, and PRISMA 2020 systematic review with optional meta-analysis (RoB 2, ROBINS-I, GRADE). - Citation Integrity: Three-index cross-validation (Semantic Scholar, OpenAlex, Crossref) plus contamination signals to detect hallucinated or post-LLM-inflection preprint sources. - Use Case: A graduate student with only a vague interest in higher education quality assurance activates Socratic mode to converge on a FINER-scored research question, then runs full mode to receive a verified, APA 7.0-formatted research report ready for handoff to paper writing. ## Quick Start Ask the agent to research a topic such as the impact of AI on higher education quality assurance, or request guided Socratic research if your question is still vague.

Frequently Asked Questions about deep-research

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

FAQPage Schema
How do I run a deep research pipeline on a topic?

Provide a research topic or question and the full mode runs six phases: scoping, investigation, analysis, composition, review, and revision. You confirm the research question and methodology after Phase 1, then receive a complete APA 7.0 report with verified citations.

What is the difference between full, quick, and systematic-review modes?

Full mode produces a 3,000-8,000 word APA report using all core agents. Quick mode delivers a 500-1,500 word brief. Systematic-review mode follows PRISMA 2020 with risk-of-bias assessment (RoB 2, ROBINS-I), meta-analysis, and GRADE tables.

How does the skill prevent hallucinated citations?

Every reference is verified against three bibliographic indexes: Semantic Scholar, OpenAlex, and Crossref. Entries also carry contamination signals flagging post-2024 preprints and unmatched sources, and unverifiable references are treated as failures rather than uncertain.

When should I use Socratic mode instead of full research mode?

Use Socratic mode when you have a vague interest but no clear research question. A mentor agent guides you through five questioning layers to converge on a FINER-scored question, and it never gives direct answers.

Can the research output be handed off to paper writing?

Yes. The research question brief, methodology blueprint, annotated bibliography, and synthesis report transfer to the academic-paper skill, whose intake agent detects these materials and skips redundant scoping and literature search steps.

What happens when the Devil's Advocate finds a critical issue?

Critical-severity findings block pipeline progression until resolved. Revision loops are capped at two iterations, and any remaining issues are documented in an acknowledged limitations section of the final report.