ybkk-AIOS
Run the following command in DeepSeek Harness:
dsh plugin install 01men/ybkk-AIOS
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install 01men/ybkk-AIOS in your DeepSeek Harness terminal to install the plugin; source code is available at https://github.com/01men/ybkk-AIOS .
About this plugin
In most enterprises, AI assets are scattered across teams and systems: MCP services run in silos, Skills and Agents lack a unified registry and permission model, AI applications wire up their own auth stacks, and no one can tell you where tokens actually go. ybkk-AIOS (RongQi) ends that fragmentation by building a single enterprise AI resource platform on the DeepSeek Harness everything-is-a-plugin architecture, unifying identity, authorization, metering, and audit under one roof.
Core capabilities span the full AI stack: organizational IAM with an OIDC Provider, MCP deployment and health governance, a Skill and plugin marketplace, Agent lifecycle management, application SSO onboarding, a model gateway with tiered routing and budget circuit-breaking, transparent usage metering with monthly reports, and audit with alerting. The platform ships in three deployment forms — standalone, mounted inside the dsh runtime, or remotely proxied — and exposes 37 ops tools natively to Agents.
Best suited for mid-size and larger enterprises running multiple AI applications or MCP services, DevOps and platform-engineering teams that need private deployment of a unified AI asset registry, and AI engineers who want Agents to participate directly in platform operations.
Use Cases
- Unified management of MCP services, Skills, Agents, and model gateways
- Standardized SSO onboarding for in-house AI applications
- Transparent usage metering, multi-tenant isolation, and audit compliance
Best For
- Mid-size and enterprise teams running multiple AI apps or MCP services
- DevOps teams needing private deployment of a unified AI asset registry
- AI engineers who want Agents to drive platform operations
Related Plugins
A network security red-team benchmark plugin for DeepSeek-V4.1/Flash featuring dual-layer kernel injection and zero-tool architecture for controlled model safety assessment.
A security red-team prompt plugin for DeepSeek-V4 that uses code-formal mapping and zero-trigger system prompts to enable unconditional output, intended for authorized testing and research only.
Adds an Auto approval mode on the official workspace-write sandbox, classifying semantic risks, asking when ambiguous, and denying destructive operations.
Undo/rollback system for DSH: config & plugin snapshots, one-click undo/redo/restore, message-level revert, secret masking, safe mode, and offline WebUI/GUI/CLI rescue tools for when DSH won't start.