Aicontainer — Development skill for Claude Code
Sandboxed devcontainer for running Claude Code, Codex, and OpenCode in bypass / auto-approve mode.
How to install Aicontainer
This entry records only its repository, not the path inside it, so there is no
exact command to give. Open stefanoginella/aicontainer and copy the folder into
~/.claude/skills/, or the file into ~/.claude/agents/.
What Aicontainer does
Sandboxed devcontainer for running Claude Code, Codex, and OpenCode in bypass / auto-approve mode.
Alternatives in Development
- Bypass 403 — Probe a 403/401 endpoint with the most-paid bypass tricks (header injection, path encoding, method swap, WAF f 4.5k ★
- Discipline — Toggle discipline mode — auto-invoke verification, brainstorming-before-coding, and review checks 2.8k ★
- 12 Yolo Auto Mode Classifier — YOLO/Auto-Mode Classifier System 2.3k ★
README
aicontainer
[](https://www.npmjs.com/package/aicontainer) [](https://github.com/stefanoginella/aicontainer/actions/workflows/release.yml) [](https://github.com/stefanoginella/aicontainer/actions/workflows/rebuild.yml) [](LICENSE) [](https://github.com/stefanoginella/aicontainer/pkgs/container/aicontainer) [](https://containers.dev/)
A sandboxed devcontainer for running [Claude Code](https://claude.ai/code), [Codex](https://github.com/openai/codex), and [OpenCode](https://opencode.ai) in bypass / auto-approve mode safely across multiple projects.
**Why?** Auto-approve is the only way these CLIs actually fly — but pointed at your real `$HOME` it also lets a prompt-injected dependency read `.env`, exfiltrate shell history, or push through your `gh` token. `aicontainer` puts the AI behind a devcontainer boundary so you can keep auto-approve on without rebuilding your machine each time.
**What you get:** filesystem isolation, **no Docker object access by default** through a digest-pinned [socket proxy](https://github.com/Tecnativa/docker-socket-proxy), a minimal shared tool hook, an always-on cloud-metadata/link-local block, and an **opt-in** outbound allowlist. Host preferences are reduced to fixed safe fields before the agent sees them; logins still carry across projects. Defaults are listed in [What's in the box](#whats-in-the-box) so you know exactly what you're adopting.
Adjacent work: same shape as the [Trail of Bits devcontainer](https://github.com/trailof
Related Skills
Ashlr Allow
Auto-approve every ashlr MCP tool so Claude Code stops prompting in bypass-permissions mode.
Auto Mode For Paseo
Paseo plugin that routes each message to a persona on Codex, Claude, OpenCode, or any other installed provider
YOLO/Auto-Mode 분류기 시스템
관찰 위치: Claude Code 내부 아키텍처 Tool 호출을 자동 승인할지 차단할지 결정하는 2단계 보안 분류 시스템입니다. Claude Code의 "auto-approve" 모드 핵심입니다.
Netvein
Local-first Chrome CDP reverse engineering & wire-level traffic MCP server, for Claude Code, Cursor, Codex, Op
Hommies
Hommie, a floating Omarchy companion that watches your coding agents (Claude Code, Codex, OpenCode, Omacode) a
SwarmBar
SwarmBar: a macOS menu bar app showing live status of CLI coding agents (Claude Code, Codex, Kimi Code, BearCo
Related Agents
Codex Agent
Runs OpenAI Codex on the user's ChatGPT subscription. Three modes - consult (OS-sandboxed read-only analysis,
Agent Of Empires
Claude Code, OpenCode, Mistral Vibe, Codex CLI, Gemini CLI Coding Agent Terminal Session manager via tmux and
Skills Manager (AI 技能管理器)
一款用于管理 AI 编程助手技能(Skills)的统一桌面应用。 无缝组织、同步和共享 Claude Code、Codex、Opencode 及其他 AI 工具的技能。 Version Platform