Consensus Pipeline
fangqian616/consensus-pipeline · skills
Multi-agent department framework for long-form complex tasks, fighting AI hallucination, validated on academic research.
★ 121GitHub stars
13Forks
2026-09-05Last updated
PythonLanguage
MITLicense
Key features
- Consensus Pipeline takes a research topic and produces a structured literature review through multi-agent debate.
- Some UI labels are bilingual (Chinese/English mix) in English mode
- Verified — the abstract directly supports the claim
- Drop files into fulltext_papers/, then run --rerun-67 to regenerate just the report + verification stage
- The pipeline in one sentenceSearch papers → 3-layer QC filter → 11 departments debate each claim → cross-department validation → generate report with confidence scores.
Requirements
- The pipeline starts by interviewing you — an AI agent asks clarifying questions to understand your research scope, constraints, and goals.
- Paper retrieval uses free open APIs (arXiv / Semantic Scholar / OpenAlex).
Install command
dsh plugin --profile web add github:fangqian616/consensus-pipeline