academic-pipeline

Orchestrates a 10-stage academic research workflow from literature research to final manuscript.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Coordinating research, writing, integrity verification, peer review, and revision across separate tools often leads to skipped verification steps, lost state between sessions, and inconsistent quality. This Skill orchestrates the full academic pipeline as a single state-tracked workflow with mandatory checkpoints and integrity gates. ## Core Features & Use Cases - 10-Stage Orchestration: Coordinates deep-research, academic-paper, and academic-paper-reviewer skills through research, writing, integrity checks, two-stage review, revision, and finalization. - Mandatory Integrity Gates: Stage 2.5 and 4.5 run 100% reference verification plus a 7-mode AI research failure checklist that blocks the pipeline on suspected hallucinated citations, fabricated results, or methodology fabrication. - State Tracking & Resume: Material Passport records pipeline state, enabling cross-session resume via resume_from_passport with boundary hashes. - Use Case: A researcher with reviewer comments on a submitted paper enters at Stage 4 (REVISE), gets a structured revision with response-to-reviewers, passes re-review and final integrity verification, and receives a formatted final manuscript with a bilingual process record PDF. ## Quick Start Ask the assistant to run the full academic pipeline for a research paper on your chosen topic, starting from the research stage.

Frequently Asked Questions about academic-pipeline

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

FAQPage Schema
How do I run a full research-to-paper workflow with AI?

Invoke the academic pipeline with your research topic and it starts at Stage 1 (RESEARCH), then dispatches writing, integrity verification, review, revision, and finalization stages in order. Each stage completion requires your confirmation before proceeding.

Can I enter the academic pipeline mid-workflow with an existing draft?

Yes, the orchestrator detects your materials and selects the entry stage: a draft enters at Stage 2.5 integrity check, reviewer comments enter at Stage 4 revision. Mid-entry cannot skip the mandatory integrity verification before review.

How does the pipeline detect hallucinated citations and fabricated results?

Stages 2.5 and 4.5 run 100% reference verification plus a 7-mode AI research failure checklist covering implementation bugs, hallucinated citations and results, shortcut reliance, methodology fabrication, and frame-lock. Suspected failures block the pipeline until you acknowledge or fix them.

Can I resume the pipeline in a new session after context reset?

Yes, set ARS_PASSPORT_RESET=1 to write boundary checkpoints to the Material Passport, then invoke resume_from_passport=<hash> in a fresh session. The orchestrator restores the recorded stage and continues from there.

When should I use a single skill instead of the full pipeline?

Use deep-research alone for literature reviews, academic-paper alone for writing without a research phase, or academic-paper-reviewer alone for a single review. The pipeline is opt-in and intended only for end-to-end workflows.