claude-docker: Instructions file for Claude Code

CLAUDE.md

claude-docker CLAUDE.md is an instructions file for Claude Code from hrubymar10/claude-docker. It costs 3,909 tokens per session, scanned A, original, MIT.

Repository-specific instructions for Claude Docker, a Docker-based sandbox that runs Claude Code inside an isolated Linux container. The project includes commands for starting, stopping, rebuilding, and entering that container.

In plain words
What is it for?
Setting up the container, running Claude Code in it, using mounted host resources, and working with its shell and Visual Studio Code integration.
Why use it?
It explains how to use the sandbox and documents safeguards such as blocking pushes to protected branches and tag pushes.

Instructions file for Claude Code

Written for Claude Code: the file is CLAUDE.md. Also seen: reads .claude/ paths; mentions CLAUDE.md; mentions Claude Code.

This is hrubymar10/claude-docker's own configuration. It tells Claude Code how to work on claude-docker 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 claude-docker configures →

Reuse

Borrowing it

Nothing to install: this file belongs to hrubymar10/claude-docker. 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/hrubymar10/claude-docker/master/CLAUDE.md
Clone the repo
git clone --depth 1 https://github.com/hrubymar10/claude-docker

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 claude-docker CLAUDE.md

README.md
[![agentmods](https://agentmods.dev/badge/instructions/hrubymar10/claude-docker/claude-md.svg)](https://agentmods.dev/instructions/hrubymar10/claude-docker/claude-md)
Your own site
<a href="https://agentmods.dev/instructions/hrubymar10/claude-docker/claude-md"><img src="https://agentmods.dev/badge/instructions/hrubymar10/claude-docker/claude-md.svg" alt="Measured on agentmods" height="20"></a>
Per session 3,909 This file is loaded in full into every session.
When invoked 3,909 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.03909 $0.03909
Opus 5 $0.01954 $0.01954
Sonnet 5 $0.00782 $0.00782
Haiku 4.5 $0.00391 $0.00391

Measured 7d ago against content hash 675e46bce01e, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-07, from the pricing page.

Security

Grade A, and why

claude-docker 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 7d 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 · 205 lines

How it starts

The opening of the file, as written. The whole thing — 205 lines — stays where its author put it; the contents beside it link to each section on GitHub.

claude-docker

Docker sandbox for running Claude Code in an isolated Linux container that mirrors the host environment.

Quick Start

bin/claude-docker-ctrl start    # build image, start container
bin/claude-docker-ctrl stop     # stop container
bin/claude-docker-ctrl status   # show container status
bin/claude-docker-ctrl shell    # shell into the container (auto-detects from host $SHELL)
bin/claude-docker-ctrl exec     # interactive Claude session in container
bin/claude-docker-ctrl rebuild  # rebuild image from scratch + restart

Project Structure

  • Dockerfile — Alpine 3.24 image
  • docker-compose.yml — base container config (auth, socket proxy, filter proxy, Go cache)
  • scripts/ — shell scripts copied into the container at build time:
    • entrypoint.sh — runtime setup: socket proxy wait, git credentials, GPG import, user drop
    • git-wrapper.sh — blocks git push to protected branches and any git push that includes tags (--tags, --follow-tags, --mirror, refs/tags/* refspecs, <remote> tag <name> shorthand). Replaces /usr/bin/git. Defense-in-depth only: git-real is still readable+executable by the unprivileged user, so a determined caller can invoke /usr/libexec/git-real/git directly. Real branch/tag protection must come from the upstream (server-side hook).
    • docker-wrapper.sh — allowlists safe docker subcommands, including build and buildx, and warns (without blocking) when a build tag would overwrite one of the sibling sandbox images. Replaces /usr/bin/docker; direct calls to the real binary still pass through the filtering proxy.
    • claude-session.sh — process-group wrapper that ensures claude + children (gopls) are killed on disconnect
    • go-install.sh — Dockerfile helper to download Go by version
  • docker-filter-proxy/ — Go reverse proxy that blocks privileged containers, host namespacing, dangerous capabilities
  • bin/claude-docker — interactive Claude session in container (-it)
  • bin/claude-docker-vscode-wrapper — VSCode claudeProcessWrapper script (-i only, no TTY)
  • bin/claude-docker-jetbrains-wrapper — JetBrains (GoLand/IntelliJ) Claude command wrapper (auto-detects TTY)
  • bin/claude-docker-ctrl — container lifecycle management
  • bin/lib/session-cleanup.sh — host-side session lifecycle shared by the bin/ wrappers (and the user's shell function): a detached watchdog HUPs the in-container session when the launching shell dies (double-forks before setsid so interactive-zsh job control can't kill it with the terminal; tracks parent identity as PID + start time to survive PID reuse), reap_stale_sessions sweeps orphaned sessions whose host client is gone, and when the last session ends the transient claude daemon is stopped so armed background tasks can't keep re-invoking Claude unattended
  • config/ — user configuration (gitignored copies + examples):
    • docker-compose.local.example.yml — template for project volume mounts
    • claude-notifier.example — template for notification script
    • .env.example — all configurable env vars
    • claude-settings.example.json — example Claude Code hooks
    • CLAUDE.md.example — example CLAUDE.md with notifier usage
  • ../aws-ai-proxy/ - optional independently running AWS credential proxy consumed when AWS_AI_PROXY_ENABLED is true (https://github.com/hrubymar10/aws-ai-proxy)
  • beeper/ — simple Go HTTP server that plays a beep sound on the host (optional)
  • gpg-keys/ — drop GPG private keys here for commit signing (gitignored)

Read the full file on GitHub · 205 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. 7d ago First seen · 205 lines · 3,909 tokens per session scan A 675e46bce01e

Subscribe to this mod's changes

claude-docker CLAUDE.md is an instructions file published in the GitHub repository hrubymar10/claude-docker (2 stars, last pushed 13d ago), licensed MIT. It adds 3,909 tokens to every session, about $0.0195 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 2026-08-31.

Related

Other instructions, from other repositories

codex AGENTS.md

AGENTS.md instructions for openai/codex, covering rust/codex-rs, the codex-core crate, code review rules, crate api surface and model visible context.

openai/codex · 5,182 tokens

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

next.js AGENTS.md

AGENTS.md instructions for vercel/next.js, covering next.js development guide, codebase structure, monorepo overview, core package: packages/next and other important packages.

vercel/next.js · 7,296 tokens

langchain AGENTS.md

AGENTS.md instructions for langchain-ai/langchain, covering global development guidelines for the langchain monorepo, corridor security analysis, project architecture and context, monorepo structure and development tools & commands.

langchain-ai/langchain · 4,469 tokens

vscode oss-third-party-notices.instructions.md

Instructions for microsoft/vscode, covering vs code oss third-party-notices pipeline, architecture, pipeline flow in ci, applying the notice (cutover) and fallback chain (never fail the build).

microsoft/vscode · 5,001 tokens

spec-kit AGENTS.md

AGENTS.md instructions for github/spec-kit, covering agents.md, about spec kit and specify, quickstart — add a new integration in 5 steps, integration architecture and integrationmanifest — file tracking.

github/spec-kit · 7,104 tokens