AI Decision System
Replayable, offline-verifiable records for AI-assisted go/no-go decisions: experiment promotions, typed model answers and human-approved actions.
翻译准备中
可自托管的 AI 软件代理平台,具备沙箱工具执行和多租户治理
AI 翻译
Self-hostable AI agent platform with sandboxed tool execution and multi-tenant governance
AI 摘要 · README
开发工具Python · HTMLGitHub @15838328874自行安装运营者未确认
试用过了吗? 点一下就好,无需登录。
尚未检测到更新。GitHub 发布版本后会显示在这里。
个人 GitHub 账号下公开仓库的所有者可以登录并验证所有权,自行管理介绍,无需管理员批准。
登录并生成验证文件,在 30 分钟内将其提交到此仓库的默认分支。其他分支或分叉仓库中的文件无效。然后在下方检查文件,即可无需管理员批准直接管理项目。请保留文件以维持管理权限。
如需登录,将前往登录页面。登录后请回到此处,再次点击所选的验证按钮。
安装型项目无需部署域名也可验证。个人仓库可通过 GitHub 登录验证,组织仓库可使用上方的仓库验证文件。
正在加载完整 README。 以下为已保存的摘录。
A self-hostable, extensible AI agent platform: sandboxed tool execution, multi-tenant
governance (JWT/quotas/rate limits/audit), a canonical trajectory log, an eval harness with
scoring and A/B, an RL data flywheel (rollout→reward→SFT/RLVR JSONL), and protocol-level
tool extensibility via MCP + Skills. Coding is the first fully-developed scenario, not the
boundary. Python reimplementation of the pi coding agent
harness, shipped as a multi-user HTTP service: streaming LLM layer, asyncio agent loop, coding
tools, and sandboxing (CubeSandbox microVMs in production, Docker locally).查看完整 README ↗目前无法加载完整 README。请通过原文链接阅读。 以下为已保存的摘录。
A self-hostable, extensible AI agent platform: sandboxed tool execution, multi-tenant
governance (JWT/quotas/rate limits/audit), a canonical trajectory log, an eval harness with
scoring and A/B, an RL data flywheel (rollout→reward→SFT/RLVR JSONL), and protocol-level
tool extensibility via MCP + Skills. Coding is the first fully-developed scenario, not the
boundary. Python reimplementation of the pi coding agent
harness, shipped as a multi-user HTTP service: streaming LLM layer, asyncio agent loop, coding
tools, and sandboxing (CubeSandbox microVMs in production, Docker locally).查看完整 README ↗