10-Stage Academic Research Pipeline for Claude Code
LinkedIn post by Charly Wargnier (DevRel; ex-Streamlit/Snowflake/Samsung) — view post
Someone dropped a full 10-stage academic research pipeline as a Claude Code plugin, built by Edward Cheng-I Wu, free and open-source.
Key point: it doesn’t write your paper for you — it hunts references, formats citations, verifies data, and runs a “devil’s advocate” agent that attacks your own thesis before a reviewer does.
Why it matters
- → Anti-AI Voice — learns your specific writing style.
- → Integrity Gates — actively hunts fabricated citations and statistical errors.
- → Simulated Peer Review — runs drafts through a 7-agent panel including the Devil’s Advocate.
- → Cheap — a full 15k-word paper costs ~
4–6 in API credits.
Install in 30 s: /plugin install academic-research-skills · repo: https://lnkd.in/e55TxjvJ
Comments worth noting
- Kurt Wellington + Kévin Ifrah: the devil’s-advocate agent is the real differentiator — most citation tools fix formatting; this one tries to find where you’re wrong first. “The expensive skill to automate.”
- Reno Filla pushes back: automating what makes research research risks accelerating paper mills — “the worst development for science”; worries about “a lost generation” of researchers.
- Bishal Shrestha middle ground: novelty and contribution still persist in researchers’ judgement of how to leverage AI.