Claude Code plugin suite covering research-to-publication pipeline: 13-agent deep research, 12-agent paper writing, 7-agent peer review, integrity gates, ~$4-6 for a 15k-word paper.
Key Takeaways
Install via /plugin marketplace add Imbad0202/academic-research-skills; requires Claude Code v3.7.0+, ANTHROPIC_API_KEY; optional Pandoc/tectonic for DOCX/PDF output.
Two integrity gate stages (2.5 and 4.5) run a 7-mode blocking checklist; a real pipeline run caught 15 fabricated references and 3 statistical errors.
Post-publication audit on the showcase paper found 21 of 68 issues missed after 3 rounds of integrity checks, exposing real limits of automated verification.
Style Calibration learns voice from past work; Writing Quality Check targets machine-prose patterns; neither goal is hiding AI use.
Companion Experiment Agent bridges Stage 1 (research) and Stage 2 (write) for code or human-study experiments, with 11-type statistical fallacy detection.
Hacker News Comment Review
The single comment questions whether unvalidated Claude Code skill repos represent a new category of low-signal noise, coining the term “skill spam” – no performance benchmarks or external validation are provided in the source.
