DG-Claw: Instructions file for Claude Code

CLAUDE.md

DG-Claw CLAUDE.md is an instructions file for Claude Code from DGatoM/DG-Claw. It costs 1,696 tokens per session, scanned B, original, MIT.

A set of Claude Code instructions for working on the DG Claw software repository. It documents the web panel, administration tools, agent containers, installation files, tests, and other project folders.

In plain words
What is it for?
Use it when developing or maintaining DG Claw, especially when changing its installer, command-line tools, web interface, agent containers, documentation, or automated tests.
Why use it?
It gives a coding agent the project structure and working rules, reducing the chance of editing the wrong component or missing the repository's testing and installation practices.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: mentions Claude Code; mentions Codex.

This is DGatoM/DG-Claw's own configuration. It tells Claude Code how to work on DG-Claw itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything DG-Claw configures →

Reuse

Borrowing it

Nothing to install: this file belongs to DGatoM/DG-Claw. 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/DGatoM/DG-Claw/main/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/DGatoM/DG-Claw

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.

agentmods badge for DG-Claw CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/dgatom/dg-claw/claude-md/github.svg)](https://agentmods.dev/instructions/dgatom/dg-claw/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/dgatom/dg-claw/claude-md"><img src="https://agentmods.dev/badge/instructions/dgatom/dg-claw/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.

agentmods 80×15 button for DG-Claw CLAUDE.md

Your own site · 80×15
<a href="https://agentmods.dev/instructions/dgatom/dg-claw/claude-md"><img src="https://agentmods.dev/badge/instructions/dgatom/dg-claw/claude-md.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 1,696 This file is loaded in full into every session.
When invoked 1,696 The same file — it is already loaded in full.
Security scan B 1 finding. 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.01696 $0.01696
Opus 5.5 $0.00678 $0.00678
Sonnet 5.5 $0.00339 $0.00339
Haiku 4.5 $0.00170 $0.00170

Measured 17d ago against content hash e338e8d7b4bc, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade B, and why

DG-Claw CLAUDE.md scanned grade B with 1 finding 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 17d 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.

Asks for rootmediumPrivilege escalation

A mod that escalates privileges can change anything on the machine, not only the project.

rodar e instalar ali mesmo, com `sudo ./install.sh`. A pasta clonada na VPS é, portanto, a
CLAUDE.md · 90 lines

How it starts

The opening of the file, as written. The whole thing — 90 lines — stays where its author put it; the contents beside it link to each section on GitHub.

DG Claw: estrutura do repositório

O DG Claw instala um painel web com chat, um Admin e a infraestrutura para agentes de trabalho em containers separados. O README apresenta o produto e o INSTALL.md contém o procedimento de instalação.

Componentes

Caminho Conteúdo
install.sh, lib/ Instalador e funções de administração
bin/dgclaw CLI de administração dos agentes
runner/ Execução de sessões Claude Code e Codex
web/ Interface web, autenticação e chat
panel/ Configurações e recursos de cada agente
container/ Imagem dos agentes
templates/ Arquivos iniciais do Admin e dos agentes
skills/ Recursos e roteiros dos agentes
docs/ Documentação do produto
tests/ Testes automatizados

A arquitetura descreve o funcionamento e os limites entre os componentes. tests/rodar.sh executa as baterias automatizadas. Os testes do instalador usam uma raiz temporária (DGCLAW_RAIZ) e substitutos dos comandos do sistema.

Contexto de uso

  • Onde o repositório é clonado. O caminho normal é clonar o DG Claw direto na VPS em que ele vai rodar e instalar ali mesmo, com sudo ./install.sh. A pasta clonada na VPS é, portanto, a instalação de destino, e não uma cópia de desenvolvimento.
  • Quem instala. Na maioria dos casos é uma pessoa que não é programadora, seguindo uma aula ou o INSTALL.md com a ajuda do próprio Claude Code ou Codex no terminal da VPS.
  • GitHub. O repositório é público. O clone funciona com git puro, sem gh e sem login no GitHub; o gh só aparece dentro da imagem dos agentes, para quem quiser publicar código depois.
  • Por que o Admin roda como root. O Admin é a peça que instala, atualiza, cria e conserta os agentes, e isso exige mexer em serviços, containers e firewall do servidor. Ele roda como root justamente para que a pessoa resolva tudo conversando pelo painel, sem precisar voltar ao terminal. Em contrapartida, só o dono do painel fala com ele, e ele não lê e-mail, não abre links de terceiros e não tem canal externo.
  • Os agentes de trabalho não rodam como root. Os agentes que a pessoa cria (e que conversam com o mundo por WhatsApp, Telegram, e-mail e navegador) ficam em containers sem root, sem acesso ao servidor nem aos vizinhos. O SEGURANCA.md detalha essa separação.
  • Versão. A v1.0.0 é a versão publicada e em uso. Os testes de instalação completa da equipe rodam em VPS separadas, conforme o roteiro de QA; isso não muda nada para quem instala na própria VPS.

Read the full file on GitHub · 90 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.

  1. 17d ago Changed · +26 lines · +600 tokens per session e338e8d7b4bc
  2. 18d ago Changed · +6 lines · +132 tokens per session 67d6e238d74f
  3. 19d ago First seen · 58 lines · 964 tokens per session scan B 2f6f4a498727

Subscribe to this mod's changes

DG-Claw CLAUDE.md is an instructions file published in the GitHub repository DGatoM/DG-Claw (20 stars, last pushed 2d ago), licensed MIT. It adds 1,696 tokens to every session, about $0.0068 per session on Opus 5.5. A static security scan graded it B with 1 finding (asks for root). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-10-18.

Related

DG-Claw CLAUDE.md compared with instructions from other repositories

DG-Claw CLAUDE.md next to 6 instructions from other repositories
ModTypeStarsLast pushLicenceScanTokensWorks with
DG-Claw CLAUDE.mdDGatoM/DG-ClawInstructions file202d agoMITB1,696Claude Code
vscode buildNext.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA9,719GitHub Copilot
next.js AGENTS.mdvercel/next.jsInstructions file143kyesterdayMITA7,461Codex, OpenCode
langchain AGENTS.mdlangchain-ai/langchainInstructions file149kyesterdayMITA4,680Codex, OpenCode
vscode oss-third-party-notices.instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA5,516GitHub Copilot
vscode copilot-instructions.mdmicrosoft/vscodeInstructions file195kyesterdayMITA3,093GitHub Copilot
deepseek-harness AGENTS.mddeepseek-ai/deepseek-harnessInstructions file274k2d agoMITA4,313Codex, OpenCode