claude-code-my-workflow is a forkable setup for using Claude Code to produce and review academic papers, slides, data analyses, and replication packages. Researchers use its agents, skills, rules, hooks, and quality checks to coordinate these tasks and verify their results. The catalogue entries define the reusable workflow components for Claude Code.
Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/pedrohcgs/claude-code-my-workflow/claude-mdgit clone --depth 1 https://github.com/pedrohcgs/claude-code-my-workflowWrote 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.
[](https://agentmods.dev/instructions/pedrohcgs/claude-code-my-workflow/claude-md)<a href="https://agentmods.dev/instructions/pedrohcgs/claude-code-my-workflow/claude-md"><img src="https://agentmods.dev/badge/instructions/pedrohcgs/claude-code-my-workflow/claude-md.svg" alt="Measured on agentmods" height="20"></a>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 | $0.02238 | $0.02238 |
| Opus 5 | $0.01119 | $0.01119 |
| Sonnet 5 | $0.00448 | $0.00448 |
| Haiku 4.5 | $0.00224 | $0.00224 |
Grade A, and why
claude-code-my-workflow 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 2d 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.
How it starts
The opening of the file, as written. The whole thing — 165 lines — stays where its author put it; the contents beside it link to each section on GitHub.
CLAUDE.MD -- Academic Project Development with Claude Code
Project: [YOUR PROJECT NAME] Institution: [YOUR INSTITUTION] Branch: main
Scope Discipline
Do exactly what was asked — nothing adjacent. Do not add README files, build scripts,
.gitignore edits, helper utilities, or extra tooling that was not requested. If an addition
looks valuable, list it as a suggestion at the end and let the user decide.
A request for a codebook is a request for a codebook. Delivering a codebook plus a README plus build scripts plus gitignore edits means the user now has to review four things to accept one, and the usual outcome is that all four get thrown away.
Before adding anything not named in the request, ask. One line is cheaper than a revert.
Core Principles
- Plan first -- enter plan mode before non-trivial tasks; save plans to
quality_reports/plans/ - Verify after -- compile/render and confirm output at the end of every task
- Single source of truth -- Beamer
.texis authoritative; Quarto.qmdderives from it - Quality gates -- nothing ships below 80/100
- [LEARN] tags -- when corrected, save
[LEARN:category] wrong → rightto MEMORY.md
Cross-session context lives in MEMORY.md; past plans, specs, and session logs are in quality_reports/.
How we verify — the references and rules that carry the verification discipline:
verification-ladder.md— the seven rungs, from qualify the checker to the external oracle, and how the review loop converges.external-oracle-process.md— running an independent frontier-model referee (Claude Code → GPT-5.6 Sol Pro) and adjudicating what it returns.provenance-and-ground-truth.md— naming and pinning your oracles, classifying divergence, and the clean-room boundary.review-fencing.md— reviewer independence is a property of the environment, not an instruction: a neutral copy outside the checkout, prior verdicts withheld, and the answer keys the repo already commits fenced off.release-engineering.md— shipping research software: message and silent-resolution censuses, frozen feature matrices for ports, hash-claimed inherited tests, and downstream consumers pinned by commit SHA.
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.
- 2d ago First seen · 165 lines · 2,238 tokens per session scan A 98855aa3b751
claude-code-my-workflow CLAUDE.md is an instructions file published in the GitHub repository pedrohcgs/claude-code-my-workflow (1,562 stars, last pushed 12d ago), licensed MIT. It adds 2,238 tokens to every session, about $0.0112 per session on Opus 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 2026-09-03.
Other instructions, from other repositories
finai-research CLAUDE.md
Claude Code instructions for csmar432/finai-research, covering 论文-研报工作流 · finresearch agent, 一句话, 快速开始, 写作轨 and 实证轨（现代 did）.
finai-research AGENTS.md
AGENTS.md instructions for csmar432/finai-research, covering agents.md — codex / copilot / claude code entry point, project: 论文-研报工作流 · finresearch agent, quick start (3 commands), 1. install (one of:) and 2. configure llm (deepseek is free for chinese use).
finai-research copilot-instructions.md
Copilot instructions for csmar432/finai-research, covering copilot instructions for 论文-研报工作流, 项目概述, 自动启动流程（每次对话必须执行）, 诊断交互（三平台统一） and 项目结构.
statsclaw CLAUDE.md
Instructions for statsclaw/statsclaw, a project described as: A workflow framework for statistical package development.
economics-paper-reviewer CLAUDE.md
Instructions for viplee110/economics-paper-reviewer: This repository is a portable English-only economics paper pre-review skill.
causal-powers AGENTS.md
Instructions for lancegui/causal-powers, a project described as: Economist's discipline for data analysis, causal inference, and econometrics — reduced-form and structural. A Claude Code plugin whose name is a homage to superpowers; it makes the silent failures of data work loud.