SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Run style calibration from context/ papers (Step 0c). Must be run before any section drafting begins, and approved by the human before proceeding.
Human-in-the-loop quality-control framework for AI-assisted scientific manuscript preparation.
This repository also configures its own agents. See what Scientific-Writing-Skills-Claude-Code-Codex tells them →
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Run style calibration from context/ papers (Step 0c). Must be run before any section drafting begins, and approved by the human before proceeding.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Run the manuscript integrity checker as Step 0a. This command must be run before any drafting session, and again after any content changes.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Draft or improve the Discussion section using: argument flow review → claim calibration → reviewer perspective.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Comprehensive quality control for all figures and tables: QC → captions → visual consistency → table design.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Run the revision quality gate on the current section or full manuscript. Used both after individual sections and as the final pre-submission check.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Draft or improve the Introduction section using the three-skill chain: benchmark positioning → section drafting → claim calibration.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Draft or improve the Methods section using: methods writing → claim evidence verification.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Draft or improve the Results section using: results writing → figure QC → claim calibration.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Run a 4-role peer reviewer simulation across the complete manuscript. Each role is a bounded, read-only subagent. All reports are presented to the human before any revisions are considered.
SFETNI/Scientific-Writing-Skills-Claude-Code-Codex
Command Claude Code
Draft and QC the Supplementary Materials: scope approval → SM writing → figure QC for SM figures.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: