by SebastianElvis · Codex Skill · ★ 65
ClawPier AI agents have OS-level access to your machine. ClawPier runs them in a sandbox. A native desktop app for managing OpenClaw and Hermes AI agent instances inside Docker containers — sandboxed from your host by default. Available on macOS, Linux, and Windows. ▶ Watch full demo (60s) Why ClawPier? AI agent runtimes like OpenClaw and Hermes have OS-level access to your email, calendars, files, and messaging platforms. Running them directly on your host means a prompt injection gives an attacker your machine (CVE-2026-25253, CVSS 8.8). ClawPier fixes this by running every agent
| Stars | 65 |
| Forks | 9 |
| Language | TypeScript |
| Category | Codex Skill |
| License | MIT |
| Quality Score | 71.624796204941/100 |
| Open Issues | 14 |
| Last Updated | 2026-04-10 |
| Created | 2026-03-13 |
| Platforms | docker, node |
| Est. Tokens | ~490k |
Explore other popular codex skill tools:
clawpier is Desktop app for managing sandboxed OpenClaw and Hermes AI agent instances via Docker. Built with Tauri v2.. It is categorized as a Codex Skill with 65 GitHub stars.
clawpier is primarily written in TypeScript. It covers topics such as ai, ai-agents, automation.
You can find installation instructions and usage details in the clawpier GitHub repository at github.com/SebastianElvis/clawpier. The project has 65 stars and 9 forks, indicating an active community.
clawpier is released under the MIT license, making it free to use and modify according to the license terms.