What is dual-ai-paper-coach CLAUDE.md?

A project guide for a workflow in which Claude Code drafts and revises a scientific paper while Codex CLI reviews it in a separate process. The two AI tools communicate through files.

Running paper-drafting, independent review, revision, claim-checking, and submission-preparation workflows.

It gives each AI a clear role and defines the project’s review cycle, writing standards, and priorities.

How dengxu11111/dual-ai-paper-coach configures Claude Code

Reuse

Borrowing it

Nothing to install: this file belongs to dengxu11111/dual-ai-paper-coach. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.

Copy the file

curl -O https://raw.githubusercontent.com/dengxu11111/dual-ai-paper-coach/main/CLAUDE.md

Clone the repo

git clone --depth 1 https://github.com/dengxu11111/dual-ai-paper-coach

Made for: Claude Code.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/instructions/dengxu11111/dual-ai-paper-coach/claude-md/github.svg)](https://agentmods.dev/instructions/dengxu11111/dual-ai-paper-coach/claude-md)

Your own site

<a href="https://agentmods.dev/instructions/dengxu11111/dual-ai-paper-coach/claude-md"><img src="https://agentmods.dev/badge/instructions/dengxu11111/dual-ai-paper-coach/claude-md/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/instructions/dengxu11111/dual-ai-paper-coach/claude-md"><img src="https://agentmods.dev/badge/instructions/dengxu11111/dual-ai-paper-coach/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 1,975 This file is loaded in full into every session.

When invoked 1,975 The same file — it is already loaded in full.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.01975	$0.01975
Opus 5.5	$0.00790	$0.00790
Sonnet 5.5	$0.00395	$0.00395
Haiku 4.5	$0.00198	$0.00198

Measured 2mo ago against content hash a313da8acf10, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

dual-ai-paper-coach CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2mo ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 149 lines

How it starts

The opening of the file, as written. The whole thing — 149 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md — your working rules in this project

This file is read automatically by Claude Code every time it enters this repo.

Project goal

Build a dual-AI paper writing + reviewing workflow. Claude Code (you) drafts the paper, revises it, and produces the submission package. Codex CLI (running in an isolated subprocess) reviews it. The two communicate only through files.

Writing yardstick: Science / Nature — broad-significance framing, Nature Reporting Summary statistical completeness, citation discipline, ≤ 5 display items. Most users will actually submit to GRL / ACP / JGR or similar; this repo deliberately measures drafts against the S/N ruler. The actual rules live in prompts/writer_system.md and prompts/reviewer_system.md.

Architecture (v0.3)

User opens Claude Code in VSCode
    ↓
/draft <topic>            ← you (Claude) write the paper using prompts/
    ↓
/codex-review             ← you invoke codex exec via Bash (subprocess)
    ↓                       (Codex reviews independently, process-isolated)
/iterate                  ← you read Codex's review, revise, then re-trigger Codex
    ↓
/claim-check              ← you and Codex score the same claims with the same prompt, then compare
    ↓
/finalize                 ← you produce DOCX + cover letter + citation placeholders

The five slash commands live in .claude/commands/, each documenting its own steps, inputs, and outputs.

Your role

You are the writer agent. Codex is the independent reviewer agent (invoked via codex exec subprocess, cannot see your prompt).

You never review your own writing — reviewing is Codex's job. Crossing that line collapses the project's core "independent review" claim.

Priorities

Working > perfect — get the chain from /draft to /finalize flowing end-to-end before polishing any single step.
Real data > template fill — Results section numbers must come from my-paper/data/analysis_results.json. Never fabricate.
Cautious wording > overclaim — default to "is consistent with"; let Codex's review decide whether stronger phrasing is justified.
File interface > in-process communication — the only channel between Claude and Codex is markdown files on disk (draft.md / review_round_N.md).
S/N yardstick > field-journal leniency — even when the user targets GRL/ACP, tighten to S/N: every number paired with effect size + 95% CI + n + test name + precise p (Reporting Summary); abstract first sentence delivers broad significance; refs ≤ 50; displays ≤ 5.

Read the full file on GitHub · 149 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2mo ago First seen · 149 lines · 1,975 tokens per session scan A a313da8acf10

Subscribe to this mod's changes

dual-ai-paper-coach CLAUDE.md is an instructions file published in the GitHub repository dengxu11111/dual-ai-paper-coach (12 stars, last pushed 5mo ago), licensed MIT. It adds 1,975 tokens to every session, about $0.0079 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Scientific Writing and docs

dual-ai-paper-coach CLAUDE.md next to 6 instructions from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
dual-ai-paper-coach CLAUDE.mddengxu11111/dual-ai-paper-coach	Instructions file	12	5mo ago	MIT	A	1,975	Claude Code
vscode buildNext.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	9,719	GitHub Copilot
next.js AGENTS.mdvercel/next.js	Instructions file	143k	yesterday	MIT	A	7,461	Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchain	Instructions file	149k	yesterday	MIT	A	4,680	Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	5,516	GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscode	Instructions file	195k	yesterday	MIT	A	3,093	GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harness	Instructions file	274k	2d ago	MIT	A	4,313	Codex, OpenCode

dual-ai-paper-coach: Instructions file for Claude Code