dsh-aicc-zhunao
Run the following command in DeepSeek Harness:
dsh plugin install Cola1018/dsh-aicc-zhunao
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install Cola1018/dsh-aicc-zhunao in DeepSeek Harness to install; the full source is at https://github.com/Cola1018/dsh-aicc-zhunao.
About this plugin
In multi-agent workflows, the root session is easily pulled into low-level tool calls such as write, edit, PowerShell, and Bash, gradually losing its role as a planner and coordinator. dsh-aicc-zhunao redefines the root session around a delegation-first architecture, keeping it focused on planning, dispatching, and synthesis rather than hands-on execution.
Three components work together: main-brain-gate.js denies tool invocations at the execution layer for the root session while child agents handle bounded implementation tasks; the aicc-law skill provides portable operating rules covering task boundaries, evidence trails, and safety checks without depending on any private system or local filesystem layout; preset.yml and agent.cordis.yml handle preset registration and orchestration composition out of the box.
Suited for teams or individuals running multi-agent workflows in DeepSeek Harness who want the root session to maintain a global perspective, delegate concrete work to child agents, and enforce task scope and safety boundaries through a set of general-purpose rules.
Use Cases
- Root session gets derailed by tool calls during planning and needs execution-layer isolation
- Multi-agent collaboration lacks unified task boundaries and safety rules
- Workflows need to stay portable without depending on private systems
Best For
- Engineers building multi-agent workflows with DeepSeek Harness
- Teams that need the root session to focus on coordination while child agents execute
- Individual developers who want general rules to constrain task scope and safety boundaries
Related Plugins
A method pack that makes AI coding agents plan against your real baseline, prove completion with fresh evidence, and reduce reworks and unsafe changes.
Turns the DeepSeek Harness session into a captain that builds a durable sub-agent team, splits goals into dependency-aware tasks, and coordinates work via direct messages and a live Web UI.
Gives coding agents design judgment, letting Claude Code, Cursor, and 70+ agents generate and iterate high-quality UI, presentations, and graphics right from the terminal.
Run the Pi ecosystem's plugins on DeepSeek Harness, unmodified, via a compatibility layer that implements Pi's public extension ABI on DSH's native services.