#448 · Primary category: AI Coding Assistants
vibekit
Run Claude Code, Gemini, Codex — or any coding agent — in a clean, isolated sandbox with sensitive data redaction and observability baked in.
Project last updated:01/13/26
GitHub Stars
1.9K
Forks
247
Contributors
19
License
MIT
Why we included this project
Claude Code, Gemini CLI, and Codex are handy until they run somewhere they shouldn't. VibeKit gives those command-line agents a contained environment: it executes their output inside isolated Docker containers, filters secrets and API keys out of completions before they reach the model, and shows you what the agent is doing with live logs and traces. That's useful for teams letting AI write code on shared or production-adjacent machines, where a slip could touch local files or leak credentials. Everything runs locally and offline, with no cloud dependency, and an SDK lets you embed the same sandboxing and monitoring into your own app. If you already use these agents but want guardrails, this is a practical option rather than another agent framework.
Articles for this project
No articles for this project yet.
To suggest a topic or contribute an article, contact us.
Related projects in this category
superpowers
An agentic skills framework & software development methodology that works.
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
skills
Skills for Real Engineers. Straight from my .agents directory.
opencode
The open source coding agent.
andrej-karpathy-skills
A single CLAUDE.md file to improve Claude Code behavior, derived from Andrej Karpathy's observations on LLM coding pitfalls.