760485464760485464Communityยท9 Agent Skills Included

qyclaw

Multi-tenant platform for running isolated, auditable agent tasks

Runs multi-user agent workloads with layered memory, skill management, MCP connections, and scheduled task queues. Isolates high-risk tool execution like shell commands and file processing inside container sandboxes instead of the main runtime. Switches between deepagents and claude backends per session while keeping audit logs, approvals, and user data intact.
npx skills add 760485464/qyclaw --all -g -y

All Skills in This Repository (9)

Pure Emerald Level Indicators

Frequently Asked Questions

FAQPage Schema
How to install Qyclaw?โ–ผ

Run `npx skills add 760485464/qyclaw --all -g -y` in your terminal to install all skills in this suite globally.

What is Qyclaw used for?โ–ผ

Qyclaw is a multi-tenant agent task platform that manages users, sessions, skills, memory, and external MCP connections, with high-risk tool execution isolated in container sandboxes.

How does Qyclaw keep agent execution safe?โ–ผ

It runs risky actions like shell commands, file writes, and skill scripts inside Docker container sandboxes, while orchestration, memory, and approvals stay in the main system with full audit logs.

Can Qyclaw switch between different agent backends?โ–ผ

Yes. You can switch per session between the deepagents and claude backends while keeping conversations, memory, skills, and connections stored in the platform.

Does Qyclaw support long-term user memory?โ–ผ

Yes. It separates user long-term memory from session memory and can integrate with Hindsight for semantic recall across conversations, with configurable reflection intervals.

Related Repositories in Software Engineering

View All in Software Engineeringโ†’