dsh-omni-router
Run the following command in DeepSeek Harness:
dsh plugin install qwe225380/dsh-omni-router
Paste the following prompt into your AI chat to install this plugin:
In DeepSeek Harness, install this plugin using the command `dsh plugin install qwe225380/dsh-omni-router` with the source code at https://github.com/qwe225380/dsh-omni-router.
About this plugin
When using coding agents for software development, common issues arise: agents may target wrong code and context, drift from complex tasks, forget acceptance criteria, skip real testing, repeat failed strategies, or claim completion without actual progress. dsh-omni-router is designed to address these engineering challenges by adding task-level reliability to DeepSeek Harness.
Its core capabilities include intelligent intervention (true NOOP for simple tasks, assisted guarding for complex ones), precise context and task intelligence, evidence verification and recovery, and a reuse-first approach that leverages existing DSH skills, tools, and plugins. By defining success conditions and failure protocols, it ensures every task is completed with verifiable proof, thereby enhancing the quality of agent workflows.
dsh-omni-router is ideal for developers and teams using DeepSeek Harness who seek high-quality code delivery and automated verification of task completion. From frontend fixes to complex refactoring, it helps improve the reliability and efficiency of coding agents in any scenario where dependable task execution is critical.
Use Cases
- Ensuring reliability during complex coding tasks
- Verifying task completion and authenticity
- Automatically adjusting recovery strategies after failure
Best For
- Developers using DeepSeek Harness
- Software teams pursuing code quality
- Engineers requiring automated verification
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.